A particle, basic element of simulation; interacts with other bodies.
Whether this body has different inertia along principal axes; NewtonIntegrator makes use of this flag to call rotation integration routine for aspherical bodies, which is more expensive.
Whether this body should have Body.bound created. Note that bodies without a bound do not participate in collision detection. (In c++, use Body::isBounded/Body::setBounded)
Returns Id of chain to which the body belongs.
Id of clump this body makes part of; invalid number if not part of clump; see Body::isStandalone, Body::isClump, Body::isClumpMember properties.
Not meant to be modified directly from Python, use O.bodies.appendClumped instead.
Return dictionary of attributes.
Whether this body will be moved by forces. (In c++, use Body::isDynamic/Body::setDynamic)
Bits of various body-related flags. Do not access directly. In c++, use isDynamic/setDynamic, isBounded/setBounded, isAspherical/setAspherical. In python, use Body.dynamic, Body.bounded, Body.aspherical.
Bitmask for determining interactions.
Unique id of this body.
Return all interactions in which this body participates.
True if this body is clump itself, false otherwise.
True if this body is clump member, false otherwise.
True if this body is neither clump, nor clump member; false otherwise.
Returns step number at which the body was added to simulation.
Shorthand for Body::groupMask
Shorthand for Body::material
Returns time at which the body was added to simulation.
Update object attributes from given dictionary
Geometry of a body
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
Update object attributes from given dictionary
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
Box (cuboid) particle geometry. (Avoid using in new code, prefer Facet instead.
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Half-size of the cuboid
Whether this Shape will be highlighted when rendered.
Update object attributes from given dictionary
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
Geometry of a deformable chained cylinder, using geometry Cylinder.
Deviation of node1 orientation from node-to-node vector
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
tensile-free length, used as reference for tensile strain
Length [m]
Radius [m]
Length vector
Update object attributes from given dictionary
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
Rigid aggregate of bodies
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
Return clump members as {‘id1’:(relPos,relOri),...}
Update object attributes from given dictionary
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
Geometry of a cylinder, as Minkowski sum of line and sphere.
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
Length [m]
Radius [m]
Length vector
Update object attributes from given dictionary
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
Facet (triangular particle) geometry.
Facet’s area
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
Facet’s normal (in local coordinate system)
Update object attributes from given dictionary
Vertex positions in local coordinates.
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
GridConnection shape. Component of a grid designed to link two GridNodes. It’s highly recommended to use utils.gridConnection(...) to generate correct GridConnections.
missing doc :(
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
true if two nodes from different periods are connected.
Radius [m]
Update object attributes from given dictionary
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
GridNode shape, component of a grid. To create a Grid, place the nodes first, they will define the spacial discretisation of it. It’s highly recommended to use utils.gridNode(...) to generate correct GridNodes. Note that the GridNodes should only be in an Interaction with other GridNodes. The Sphere-Grid contact is only handled by the GridConnections.
List of GridConnections the GridNode is connected to.
Add a GridConnection to the GridNode.
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
Radius [m]
Update object attributes from given dictionary
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
Polyhedral (convex) geometry.
return polyhedra’s centroid
return polyhedra’s inertia tensor
return polyhedra’s orientation
return polyhedra’s volume
Initialization
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
Seed for random generator.
Size of the grain in meters - x,y,z - before random rotation
Update object attributes from given dictionary
Tetrahedron vertices in global coordinate system.
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
Geometry of spherical particle.
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
Radius [m]
Update object attributes from given dictionary
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
Tetrahedron geometry.
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
Update object attributes from given dictionary
Tetrahedron vertices (in local coordinate system).
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
Object representing infinite plane aligned with the coordinate system (axis-aligned wall).
Axis of the normal; can be 0,1,2 for +x, +y, +z respectively (Body’s orientation is disregarded for walls)
Color for rendering (normalized RGB).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Whether this Shape will be highlighted when rendered.
Which side of the wall interacts: -1 for negative only, 0 for both, +1 for positive only
Update object attributes from given dictionary
Whether this Shape is rendered using color surfaces, or only wireframe (can still be overridden by global config of the renderer).
State of a body (spatial configuration, internal variables).
Current angular momentum
Current angular velocity
Degress of freedom where linear/angular velocity will be always constant (equal to zero, or to an user-defined value), regardless of applied force/torque. String that may contain ‘xyzXYZ’ (translations and rotations).
(auto-updated) see GlobalStiffnessTimeStepper::targetDt.
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Displacement from reference position (pos - refPos)
Inertia of associated body, in local coordinate system.
Damping in Newtonintegrator can be deactivated for individual particles by setting this variable to FALSE. E.g. damping is inappropriate for particles in free flight under gravity but it might still be applicable to other particles in the same simulation.
Mass of this body
Current orientation.
Current position.
Reference orientation
Reference position
Rotation from reference orientation (as rotation vector)
Position and orientation as one object.
Update object attributes from given dictionary
Current linear velocity.
State of a chained bodies, containing information on connectivity in order to track contacts jumping over contiguous elements. Chains are 1D lists from which id of chained bodies are retrieved via rank and chainNumber.
Add body to current active chain
Current angular momentum
Current angular velocity
id of the body containing - for postLoad operations only.
Degress of freedom where linear/angular velocity will be always constant (equal to zero, or to an user-defined value), regardless of applied force/torque. String that may contain ‘xyzXYZ’ (translations and rotations).
chain id.
(auto-updated) see GlobalStiffnessTimeStepper::targetDt.
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Displacement from reference position (pos - refPos)
Inertia of associated body, in local coordinate system.
Damping in Newtonintegrator can be deactivated for individual particles by setting this variable to FALSE. E.g. damping is inappropriate for particles in free flight under gravity but it might still be applicable to other particles in the same simulation.
Mass of this body
Current orientation.
Current position.
rank in the chain.
Reference orientation
Reference position
Rotation from reference orientation (as rotation vector)
Position and orientation as one object.
Update object attributes from given dictionary
Current linear velocity.
State information about body use by cpm-model.
None of that is used for computation (at least not now), only for post-processing.
Current angular momentum
Current angular velocity
Degress of freedom where linear/angular velocity will be always constant (equal to zero, or to an user-defined value), regardless of applied force/torque. String that may contain ‘xyzXYZ’ (translations and rotations).
Damage tensor computed with microplane theory averaging. state.damageTensor.trace() = state.normDmg
(auto-updated) see GlobalStiffnessTimeStepper::targetDt.
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Displacement from reference position (pos - refPos)
Volumetric strain around this body (unused for now)
Inertia of associated body, in local coordinate system.
Damping in Newtonintegrator can be deactivated for individual particles by setting this variable to FALSE. E.g. damping is inappropriate for particles in free flight under gravity but it might still be applicable to other particles in the same simulation.
Mass of this body
Average damage including already deleted contacts (it is really not damage, but 1-relResidualStrength now)
Number of (cohesive) contacts that damaged completely
Number of contacts with this body
Current orientation.
Current position.
Reference orientation
Reference position
Rotation from reference orientation (as rotation vector)
Position and orientation as one object.
Stress tensor of the spherical particle (under assumption that particle volume = pi*r*r*r*4/3.) for packing fraction 0.62
Update object attributes from given dictionary
Current linear velocity.
JCFpm state information about each body.
Current angular momentum
Current angular velocity
Degress of freedom where linear/angular velocity will be always constant (equal to zero, or to an user-defined value), regardless of applied force/torque. String that may contain ‘xyzXYZ’ (translations and rotations).
(auto-updated) see GlobalStiffnessTimeStepper::targetDt.
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Displacement from reference position (pos - refPos)
Inertia of associated body, in local coordinate system.
Damping in Newtonintegrator can be deactivated for individual particles by setting this variable to FALSE. E.g. damping is inappropriate for particles in free flight under gravity but it might still be applicable to other particles in the same simulation.
Indicates the number of joint surfaces to which the particle belongs (0-> no joint, 1->1 joint, etc..). [-]
Specifies the normal direction to the joint plane 1. Rk: the ideal here would be to create a vector of vector wich size is defined by the joint integer (as much joint normals as joints). However, it needs to make the pushback function works with python since joint detection is done through a python script. lines 272 to 312 of cpp file should therefore be adapted. [-]
Specifies the normal direction to the joint plane 2. [-]
Specifies the normal direction to the joint plane 3. [-]
Mass of this body
Number of initial cohesive interactions. [-]
Identifies if the particle is on a joint surface.
Current orientation.
Current position.
Reference orientation
Reference position
Rotation from reference orientation (as rotation vector)
Position and orientation as one object.
Number of shear breakages. [-]
Relative number (in [0;1], compared with :yref:noIniLinks) of shear breakages. [-]
Number of tensile breakages. [-]
Relative number (in [0;1], compared with :yref:noIniLinks) of tensile breakages. [-]
Update object attributes from given dictionary
Current linear velocity.
Wire state information of each body.
None of that is used for computation (at least not now), only for post-processing.
Current angular momentum
Current angular velocity
Degress of freedom where linear/angular velocity will be always constant (equal to zero, or to an user-defined value), regardless of applied force/torque. String that may contain ‘xyzXYZ’ (translations and rotations).
(auto-updated) see GlobalStiffnessTimeStepper::targetDt.
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Displacement from reference position (pos - refPos)
Inertia of associated body, in local coordinate system.
Damping in Newtonintegrator can be deactivated for individual particles by setting this variable to FALSE. E.g. damping is inappropriate for particles in free flight under gravity but it might still be applicable to other particles in the same simulation.
Mass of this body
Number of broken links (e.g. number of wires connected to the body which are broken). [-]
Current orientation.
Current position.
Reference orientation
Reference position
Rotation from reference orientation (as rotation vector)
Position and orientation as one object.
Update object attributes from given dictionary
Current linear velocity.
Material properties of a body.
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Update object attributes from given dictionary
material for bubble interactions, for use with other Bubble classes
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
The surface tension in the fluid surrounding the bubbles. The default value is that of water at 25 degrees Celcius.
Update object attributes from given dictionary
Dimensionless coefficient used for the rolling stiffness.
Dimensionless coefficient used for the twist stiffness.
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Dimensionless coefficient used to calculate the plastic rolling moment (if negative, plasticity will not be applied).
Contact friction angle (in radians). Hint : use ‘radians(degreesValue)’ in python scripts.
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Use bending/twisting moment at contact. The contact will have moments only if both bodies have this flag true. See CohFrictPhys::cohesionDisablesFriction for details.
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Poisson’s ratio or the ratio between shear and normal stiffness [-]. It has different meanings depending on the Ip functor.
Update object attributes from given dictionary
elastic modulus [Pa]. It has different meanings depending on the Ip functor.
Concrete material, for use with other Cpm classes.
Note
Density is initialized to 4800 kgm⁻³automatically, which gives approximate 2800 kgm⁻³ on 0.5 density packing.
Concrete Particle Model (CPM)
CpmMat is particle material, Ip2_CpmMat_CpmMat_CpmPhys averages two particles’ materials, creating CpmPhys, which is then used in interaction resultion by Law2_ScGeom_CpmPhys_Cpm. CpmState is associated to CpmMat and keeps state defined on particles rather than interactions (such as number of completely damaged interactions).
The model is contained in externally defined macro CPM_MATERIAL_MODEL, which features damage in tension, plasticity in shear and compression and rate-dependence. For commercial reasons, rate-dependence and compression-plasticity is not present in reduced version of the model, used when CPM_MATERIAL_MODEL is not defined. The full model will be described in detail in my (Václav Šmilauer) thesis along with calibration procedures (rigidity, poisson’s ratio, compressive/tensile strength ratio, fracture energy, behavior under confinement, rate-dependent behavior).
Even the public model is useful enough to run simulation on concrete samples, such as uniaxial tension-compression test.
Law for damage evolution in uniaxial tension. 0 for linear stress-strain softening branch, 1 (default) for exponential damage evolution law
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Exponent for normal viscosity function. [-]
Characteristic time for normal viscosity. [s]
Limit elastic strain [-]
Contact friction angle (in radians). Hint : use ‘radians(degreesValue)’ in python scripts.
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Isotropic prestress of the whole specimen. [Pa]
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
If true, no damage will occur (for testing only).
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Exponent for visco-plasticity function. [-]
Characteristic time for visco-plasticity. [s]
Poisson’s ratio or the ratio between shear and normal stiffness [-]. It has different meanings depending on the Ip functor.
relative ductility of bonds in normal direction
Initial cohesion [Pa]
Update object attributes from given dictionary
elastic modulus [Pa]. It has different meanings depending on the Ip functor.
Purely elastic material. The material parameters may have different meanings depending on the IPhysFunctor used : true Young and Poisson in Ip2_FrictMat_FrictMat_MindlinPhys, or contact stiffnesses in Ip2_FrictMat_FrictMat_FrictPhys.
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Poisson’s ratio or the ratio between shear and normal stiffness [-]. It has different meanings depending on the Ip functor.
Update object attributes from given dictionary
elastic modulus [Pa]. It has different meanings depending on the Ip functor.
Elastic material with contact friction. See also ElastMat.
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Contact friction angle (in radians). Hint : use ‘radians(degreesValue)’ in python scripts.
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Poisson’s ratio or the ratio between shear and normal stiffness [-]. It has different meanings depending on the Ip functor.
Update object attributes from given dictionary
elastic modulus [Pa]. It has different meanings depending on the Ip functor.
Dimensionless coefficient used for the rolling stiffness.
Dimensionless coefficient used for the twist stiffness.
Compresion elasticity modulus
Bending creeping coefficient. Usual values between 0 and 1.
Tension/compression creeping coefficient. Usual values between 0 and 1.
Twist creeping coefficient. Usual values between 0 and 1.
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Maximal plastic strain compression
Maximal plastic strain tension
Maximal plastic bending strain
Maximal plastic twist strain
Contact friction angle (in radians). Hint : use ‘radians(degreesValue)’ in python scripts.
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Bending elastic stress limit
Twist elastic stress limit
Poisson’s ratio or the ratio between shear and normal stiffness [-]. It has different meanings depending on the Ip functor.
Shear elastic stress limit
shear elasticity modulus
Compression elastic stress limit
Tension elastic stress limit
Tension elasticity modulus
Bending plastic unload coefficient. Usual values between 0 and +infinity.
Tension/compression plastic unload coefficient. Usual values between 0 and +infinity.
Twist plastic unload coefficient. Usual values between 0 and +infinity.
Update object attributes from given dictionary
elastic modulus [Pa]. It has different meanings depending on the Ip functor.
Possibly jointed, cohesive frictional material, for use with other JCFpm classes
Defines the maximum admissible tangential force in shear, for Fn=0, in the matrix (FsMax = cohesion * crossSection). [Pa]
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Contact friction angle (in radians). Hint : use ‘radians(degreesValue)’ in python scripts.
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Defines the maximum admissible tangential force in shear, for Fn=0, on the joint surface. [Pa]
Defines the dilatancy of the joint surface (only valid for smooth contact logic). [rad]
Defines Coulomb friction on the joint surface. [rad]
Defines the normal stiffness on the joint surface. [Pa/m]
Defines the shear stiffness on the joint surface. [Pa/m]
Defines the maximum admissible normal force in traction on the joint surface. [Pa]
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Poisson’s ratio or the ratio between shear and normal stiffness [-]. It has different meanings depending on the Ip functor.
Defines the maximum admissible normal force in traction in the matrix (FnMax = tensileStrength * crossSection). [Pa]
If particles of two different types interact, it will be with friction only (no cohesion).[-]
Update object attributes from given dictionary
elastic modulus [Pa]. It has different meanings depending on the Ip functor.
Material for simple Ludning`s model of contact.
Viscous damping
Dimensionless plasticity depth
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Friction angle [rad]
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Slope of loading plastic branch
Slope of irreversible, tensile adhesive branch
Slope of unloading and reloading limit elastic branch
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Update object attributes from given dictionary
Material class for particles whose contact obey to a normal inelasticity (governed by this coeff_dech).
=kn(unload) / kn(load)
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Contact friction angle (in radians). Hint : use ‘radians(degreesValue)’ in python scripts.
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Poisson’s ratio or the ratio between shear and normal stiffness [-]. It has different meanings depending on the Ip functor.
Update object attributes from given dictionary
elastic modulus [Pa]. It has different meanings depending on the Ip functor.
Elastic material with Coulomb friction.
To be splitted ... or not
Normal volumetric ‘stiffness’ (N/m3).
Shear stiffness (N/m).
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Contact friction angle (in radians).
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Stress at whis polyhedra of volume 4/3*pi [mm] breaks.
Update object attributes from given dictionary
Material for simple viscoelastic model of contact.
Note
Shop::getViscoelasticFromSpheresInteraction (and utils.getViscoelasticFromSpheresInteraction in python) compute kn, cn, ks, cs from analytical solution of a pair spheres interaction problem.
True, if capillar forces need to be added.
Different types of capillar interaction: Willett_numeric, Willett_analytic [Willett2000] , Weigert [Weigert1999] , Rabinovich [Rabinov2005]
Liquid bridge volume [m^3]
Normal viscous constant
Shear viscous constant
Density of the material [kg/m³]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Contact friction angle (in radians). Hint : use ‘radians(degreesValue)’ in python scripts.
Surface tension [N/m]
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Normal elastic stiffness
Shear elastic stiffness
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Rolling resistance, see [Zhou1999536].
Rolling resistance type, see [Zhou1999536]. mRtype=1 - equation (3) in [Zhou1999536]; mRtype=2 - equation (4) in [Zhou1999536].
Stiffness and viscosity are multiplied by the reduced mass. If massMultiply=false, these parameter are set explicitly without mass multiplication
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Poisson’s ratio or the ratio between shear and normal stiffness [-]. It has different meanings depending on the Ip functor.
Contact angle [°]
Update object attributes from given dictionary
elastic modulus [Pa]. It has different meanings depending on the Ip functor.
Material for use with the Wire classes
Cross-section area of a single wire used to transform stress into force. [m²]
Density of the material [kg/m³]
Diameter of the single wire in [m] (the diameter is used to compute the cross-section area of the wire).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Contact friction angle (in radians). Hint : use ‘radians(degreesValue)’ in python scripts.
Numeric id of this material; is non-negative only if this Material is shared (i.e. in O.materials), -1 otherwise. This value is set automatically when the material is inserted to the simulation via O.materials.append. (This id was necessary since before boost::serialization was used, shared pointers were not tracked properly; it might disappear in the future)
Type of the mesh. If true two particles of the same material which body ids differ by one will be considered as double-twisted interaction.
Textual identifier for this material; can be used for shared materials lookup in MaterialContainer.
Parameter between 0 and 1 to reduce strain at failure of a double-twisted wire (as used by [Bertrand2008]). [-]
Parameter between 0 and 1 introduced by [Thoeni2013] which defines where the shifted force-displacement curve intersects with the new initial stiffness: . [-]
Parameter between 0 and 1 to compute the elastic stiffness of a double-twisted wire (as used by [Bertrand2008]): . [-]
Parameter between 0 and 1 introduced by [Thoeni2013] which defines the maximum shift of the force-displacement curve in order to take an additional initial elongation (e.g. wire distortion/imperfections, slipping, system flexibility) into account: . [-]
Return new State instance, which is associated with this Material. Some materials have special requirement on Body::state type and calling this function when the body is created will ensure that they match. (This is done automatically if you use utils.sphere, … functions from python).
Poisson’s ratio or the ratio between shear and normal stiffness [-]. It has different meanings depending on the Ip functor.
Integer used to initialize the random number generator for the calculation of the distortion. If the integer is equal to 0 a internal seed number based on the time is computed. [-]
Piecewise linear definition of the stress-strain curve by set of points (strain[-]>0,stress[Pa]>0) for one single wire. Tension only is considered and the point (0,0) is not needed! NOTE: Vector needs to be initialized!
Piecewise linear definition of the stress-strain curve by set of points (strain[-]>0,stress[Pa]>0) for the double twist. Tension only is considered and the point (0,0) is not needed! If this value is given the calculation will be based on two different stress-strain curves without considering the parameter introduced by [Bertrand2008] (see [Thoeni2013]).
Three different types are considered:
0 | Corresponds to Bertrand’s approach (see [Bertrand2008]): only one stress-strain curve is used |
1 | New approach: two separate stress-strain curves can be used (see [Thoeni2013]) |
2 | New approach with stochastically distorted contact model: two separate stress-strain curves with changed initial stiffness and horizontal shift (shift is random if ![]() |
By default the type is 0.
Update object attributes from given dictionary
elastic modulus [Pa]. It has different meanings depending on the Ip functor.
Object bounding part of space taken by associated body; might be larger, used to optimalize collision detection
Color for rendering this object
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
record iteration of last reference position update (auto-updated)
Reference position, updated at current body position each time the bound dispatcher update bounds (auto-updated)
The length used to increase the bounding boxe size, can be adjusted on the basis of previous displacement if BoundDispatcher::targetInterv>0. (auto-updated)
Update object attributes from given dictionary
Axis-aligned bounding box, for use with InsertionSortCollider. (This class is quasi-redundant since min,max are already contained in Bound itself. That might change at some point, though.)
Color for rendering this object
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
record iteration of last reference position update (auto-updated)
Reference position, updated at current body position each time the bound dispatcher update bounds (auto-updated)
The length used to increase the bounding boxe size, can be adjusted on the basis of previous displacement if BoundDispatcher::targetInterv>0. (auto-updated)
Update object attributes from given dictionary
Interaction between pair of bodies.
Distance of bodies in cell size units, if using periodic boundary conditions; id2 is shifted by this number of cells from its State::pos coordinates for this interaction to exist. Assigned by the collider.
Warning
(internal) cellDist must survive Interaction::reset(), it is only initialized in ctor. Interaction that was cancelled by the constitutive law, was reset() and became only potential must have thepriod information if the geometric functor again makes it real. Good to know after few days of debugging that :-)
Return dictionary of attributes.
Geometry part of the interaction.
True if this interaction is active. Otherwise the forces from this interaction will not be taken into account. True by default.
True if this interaction has both geom and phys; False otherwise.
Step number at which the interaction was added to simulation.
Step number at which the interaction was fully (in the sense of geom and phys) created. (Should be touched only by IPhysDispatcher and InteractionLoop, therefore they are made friends of Interaction
Physical (material) part of the interaction.
Update object attributes from given dictionary
Geometrical configuration of interaction
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Update object attributes from given dictionary
Test
Bending at contact as a vector defining axis of rotation and angle (angle=norm).
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Return incident velocity of the interaction (see also Ig2_Sphere_Sphere_ScGeom.avoidGranularRatcheting for explanation of the ratcheting argument).
Orientation of body 1 one at initialisation time (auto-updated)
Orientation of body 2 one at initialisation time (auto-updated)
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
Penetration distance of spheres (positive if overlapping)
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Return relative angular velocity of the interaction.
Shear displacement increment in the last step
Elastic twist angle of the contact.
Stored creep, substracted from total relative rotation for computation of elastic moment (auto-updated)
Update object attributes from given dictionary
Geometry of a cylinder-sphere contact.
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
position of 2nd node (auto-updated)
id of next chained cylinder (auto-updated)
Return incident velocity of the interaction (see also Ig2_Sphere_Sphere_ScGeom.avoidGranularRatcheting for explanation of the ratcheting argument).
this flag is turned true (1) automatically if the contact is shared between two chained cylinders. A duplicated interaction will be skipped once by the constitutive law, so that only one contact at a time is effective. If isDuplicate=2, it means one of the two duplicates has no longer geometric interaction, and should be erased by the constitutive laws.
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
contact on node?
Penetration distance of spheres (positive if overlapping)
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Return relative angular velocity of the interaction.
position of the contact on the cylinder (0: node-, 1:node+) (auto-updated)
Shear displacement increment in the last step
position of 1st node (auto-updated)
Defines the body id of the cylinder where the contact is real, when CylScGeom::isDuplicate>0.
Update object attributes from given dictionary
Class representing geometry of two bodies in contact. The contact has 6 DOFs (normal, 2×shear, twist, 2xbending) and uses ScGeom incremental algorithm for updating shear.
Bending at contact as a vector defining axis of rotation and angle (angle=norm).
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
position of 2nd node (auto-updated)
id of next chained cylinder (auto-updated)
Return incident velocity of the interaction (see also Ig2_Sphere_Sphere_ScGeom.avoidGranularRatcheting for explanation of the ratcheting argument).
Orientation of body 1 one at initialisation time (auto-updated)
Orientation of body 2 one at initialisation time (auto-updated)
this flag is turned true (1) automatically if the contact is shared between two chained cylinders. A duplicated interaction will be skipped once by the constitutive law, so that only one contact at a time is effective. If isDuplicate=2, it means one of the two duplicates has no longer geometric interaction, and should be erased by the constitutive laws.
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
contact on node?
Penetration distance of spheres (positive if overlapping)
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Return relative angular velocity of the interaction.
position of the contact on the cylinder (0: node-, 1:node+) (auto-updated)
Shear displacement increment in the last step
position of 1st node (auto-updated)
Defines the body id of the cylinder where the contact is real, when CylScGeom::isDuplicate>0.
Elastic twist angle of the contact.
Stored creep, substracted from total relative rotation for computation of elastic moment (auto-updated)
Update object attributes from given dictionary
Class uniting ScGeom and L3Geom, for the purposes of GlobalStiffnessTimeStepper. (It might be removed inthe future). Do not use this class directly.
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Update object attributes from given dictionary
Geometry of a GridConnection-GridConnection contact.
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Return incident velocity of the interaction (see also Ig2_Sphere_Sphere_ScGeom.avoidGranularRatcheting for explanation of the ratcheting argument).
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
Penetration distance of spheres (positive if overlapping)
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Return relative angular velocity of the interaction.
position of the contact on the first connection (0: node-, 1:node+) (auto-updated)
position of the contact on the first connection (0: node-, 1:node+) (auto-updated)
Shear displacement increment in the last step
Update object attributes from given dictionary
Geometry of a GridNode-GridNode contact. Inherits almost everything from ScGeom6D.
Bending at contact as a vector defining axis of rotation and angle (angle=norm).
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Return incident velocity of the interaction (see also Ig2_Sphere_Sphere_ScGeom.avoidGranularRatcheting for explanation of the ratcheting argument).
Orientation of body 1 one at initialisation time (auto-updated)
Orientation of body 2 one at initialisation time (auto-updated)
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
Penetration distance of spheres (positive if overlapping)
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Return relative angular velocity of the interaction.
Shear displacement increment in the last step
Elastic twist angle of the contact.
Stored creep, substracted from total relative rotation for computation of elastic moment (auto-updated)
Update object attributes from given dictionary
Geometry of contact given in local coordinates with 3 degress of freedom: normal and two in shear plane. [experimental]
Applied force in local coordinates [debugging only, will be removed]
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Transformation (rotation) from global to local coordinates. (the translation part is in GenericSpheresContact.contactPoint)
Displacement components, in local coordinates. (auto-updated)
Zero displacement value; u0 should be always subtracted from the geometrical displacement u computed by appropriate IGeomFunctor, resulting in u. This value can be changed for instance
Note
Never set an absolute value of u0, only increment, since both IGeomFunctor and LawFunctor use it. If you need to keep track of plastic deformation, store it in IPhys isntead (this might be changed: have u0 for LawFunctor exclusively, and a separate value stored (when that is needed) inside classes deriving from L3Geom.
Update object attributes from given dictionary
Geometric of contact in local coordinates with 6 degrees of freedom. [experimental]
Applied force in local coordinates [debugging only, will be removed]
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
Rotation components, in local coordinates. (auto-updated)
Zero rotation, should be always subtracted from phi to get the value. See L3Geom.u0.
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Transformation (rotation) from global to local coordinates. (the translation part is in GenericSpheresContact.contactPoint)
Displacement components, in local coordinates. (auto-updated)
Zero displacement value; u0 should be always subtracted from the geometrical displacement u computed by appropriate IGeomFunctor, resulting in u. This value can be changed for instance
Note
Never set an absolute value of u0, only increment, since both IGeomFunctor and LawFunctor use it. If you need to keep track of plastic deformation, store it in IPhys isntead (this might be changed: have u0 for LawFunctor exclusively, and a separate value stored (when that is needed) inside classes deriving from L3Geom.
Update object attributes from given dictionary
Geometry of interaction between 2 vector, including volumetric characteristics
Contact point (global coords), centriod of the overlapping polyhedron
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Cross-section area of the overlap (perpendicular to the normal) - not used
volume / equivalentCrossSection - not used
Normal direction of the interaction
Volume of overlap [m³]
Shear displacement increment in the last step
Update object attributes from given dictionary
Class representing geometry of a contact point between two bodies. It is more general than sphere-sphere contact even though it is primarily focused on spheres interactions (reason for the ‘Sc’ namming); it is also used for representing contacts of a Sphere with non-spherical bodies (Facet, Plane, Box, ChainedCylinder), or between two non-spherical bodies (ChainedCylinder). The contact has 3 DOFs (normal and 2×shear) and uses incremental algorithm for updating shear.
We use symbols ,
,
respectively for position, linear and angular velocities (all in global coordinates) and
for particles radii; subscripted with 1 or 2 to distinguish 2 spheres in contact. Then we define branch length and unit contact normal
The relative velocity of the spheres is then
where the fraction multplying translational velocities is to make the definition objective and avoid ratcheting effects (see Ig2_Sphere_Sphere_ScGeom.avoidGranularRatcheting). The shear component is
Tangential displacement increment over last step then reads
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Return incident velocity of the interaction (see also Ig2_Sphere_Sphere_ScGeom.avoidGranularRatcheting for explanation of the ratcheting argument).
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
Penetration distance of spheres (positive if overlapping)
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Return relative angular velocity of the interaction.
Shear displacement increment in the last step
Update object attributes from given dictionary
Class representing geometry of two bodies in contact. The contact has 6 DOFs (normal, 2×shear, twist, 2xbending) and uses ScGeom incremental algorithm for updating shear.
Bending at contact as a vector defining axis of rotation and angle (angle=norm).
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Return incident velocity of the interaction (see also Ig2_Sphere_Sphere_ScGeom.avoidGranularRatcheting for explanation of the ratcheting argument).
Orientation of body 1 one at initialisation time (auto-updated)
Orientation of body 2 one at initialisation time (auto-updated)
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
Penetration distance of spheres (positive if overlapping)
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Return relative angular velocity of the interaction.
Shear displacement increment in the last step
Elastic twist angle of the contact.
Stored creep, substracted from total relative rotation for computation of elastic moment (auto-updated)
Update object attributes from given dictionary
Geometry of a GridConnection-Sphere contact.
Bending at contact as a vector defining axis of rotation and angle (angle=norm).
some reference point for the interaction (usually in the middle). (auto-computed)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Return incident velocity of the interaction (see also Ig2_Sphere_Sphere_ScGeom.avoidGranularRatcheting for explanation of the ratcheting argument).
Orientation of body 1 one at initialisation time (auto-updated)
Orientation of body 2 one at initialisation time (auto-updated)
this flag is turned true (1) automatically if the contact is shared between two Connections. A duplicated interaction will be skipped once by the constitutive law, so that only one contact at a time is effective. If isDuplicate=2, it means one of the two duplicates has no longer geometric interaction, and should be erased by the constitutive laws.
Unit vector oriented along the interaction, from particle #1, towards particle #2. (auto-updated)
Penetration distance of spheres (positive if overlapping)
Reference radius of particle #1. (auto-computed)
Reference radius of particle #2. (auto-computed)
Return relative angular velocity of the interaction.
position of the contact on the connection (0: node-, 1:node+) (auto-updated)
Shear displacement increment in the last step
Defines the body id of the GridConnection where the contact is real, when ScGridCoGeom::isDuplicate>0.
Elastic twist angle of the contact.
Stored creep, substracted from total relative rotation for computation of elastic moment (auto-updated)
Update object attributes from given dictionary
Geometry of interaction between 2 tetrahedra, including volumetric characteristics
Contact point (global coords)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Cross-section of the overlap (perpendicular to the axis of least inertia
??
??
??
Normal of the interaction, directed in the sense of least inertia of the overlap volume
Volume of overlap [m³]
Update object attributes from given dictionary
EXPERIMENTAL. Geometry of interaction between 2 tetrahedra
Contact point (global coords)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
TODO
Normal of the interaction TODO
Volume of overlap [m³]
Update object attributes from given dictionary
Physical (material) properties of interaction.
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Update object attributes from given dictionary
Physics of bubble-bubble interactions, for use with BubbleMat
Computes the normal force acting between the two interacting bubbles using the Newton-Rhapson method
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Contact normal force
Maximum number of force iterations allowed
Convergence criteria for force iterations
Normal force
Average radius of the two interacting bubbles
Surface tension of the surrounding liquid
Update object attributes from given dictionary
Physics (of interaction) for Law2_ScGeom_CapillaryPhys_Capillarity.
Defines the surface area wetted by the meniscus on the smallest grains of radius R1 (R1<R2)
Defines the surface area wetted by the meniscus on the biggest grains of radius R2 (R1<R2)
Value of the capillary pressure Uc defines as Ugas-Uliquid
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Capillary Force produces by the presence of the meniscus
Indicates the number of meniscii that overlap with this one
If true, capillary force is zero and liquid bridge is inactive.
Normal stiffness
Shear stiffness
Presence of a meniscus if true
Normal force after previous step (in global coordinates).
Shear force after previous step (in global coordinates).
tan of angle of friction
Update object attributes from given dictionary
Volume of the menicus
is cohesion active? will be set false when a fragile contact is broken
is shear strength the sum of friction and adhesion or only adhesion?
creep viscosity [Pa.s/m].
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
do cohesion disapear when contact strength is exceeded?
Initialize the cohesive behaviour with current state as equilibrium state (same as Ip2_CohFrictMat_CohFrictMat_CohFrictPhys::setCohesionNow but acting on only one interaction)
Normal stiffness
rotational stiffness [N.m/rad]
Shear stiffness
twist stiffness [N.m/rad]
Coefficient to determine the maximum plastic rolling moment.
Maximum elastic value for the twisting moment (if zero, plasticity will not be applied). In CohFrictMat a parameter should be added to decide what value should be attributed to this threshold value.
use bending/twisting moment at contacts. See CohFrictPhys::cohesionDisablesFriction for details.
Bending moment
Twist moment
tensile strength
Normal force after previous step (in global coordinates).
cohesive part of the shear strength (a frictional term might be added depending on Law2_ScGeom6D_CohFrictPhys_CohesionMoment::always_use_moment_law)
Shear force after previous step (in global coordinates).
tan of angle of friction
plastic normal displacement, only used for tensile behaviour and if CohFrictPhys::fragile`=false. :ydefault:`0
maximum value of plastic normal displacement, after that the interaction breaks even if CohFrictPhys::fragile`=false. The default value (0) means no maximum. :ydefault:`0
Update object attributes from given dictionary
Representation of a single interaction of the Cpm type: storage for relevant parameters.
Evolution of the contact is governed by Law2_ScGeom_CpmPhys_Cpm, that includes damage effects and chages of parameters inside CpmPhys. See cpm-model for details.
normal modulus (stiffness / crossSection) [Pa]
Magnitude of normal force.
Magnitude of shear force
shear modulus [Pa]
equivalent cross-section associated with this contact [m²]
Law for softening part of uniaxial tension. 0 for linear, 1 for exponential (default)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
damage viscous overstress (at previous step or at current step)
exponent in the rate-dependent damage evolution
damage strain (at previous or current step)
characteristic time for damage (if non-positive, the law without rate-dependence is used)
strain at which the material starts to behave non-linearly
strain at which the bond is fully broken [-]
Current normal strain
normal plastic strain (initially zero)
Total shear strain (either computed from increments with ScGeom) (auto-updated)
shear plastic strain (initially zero)
Transversal strain (perpendicular to the contact axis)
Damage evolution law, evaluating the parameter.
is historically maximum strain, epsCrackOnset (
) = CpmPhys.epsCrackOnset, epsFracture = CpmPhys.epsFracture; if neverDamage is True, the value returned will always be 0 (no damage). TODO
Inversion of damage evolution law, evaluating the parameter.
is damage, for other parameters see funcG function
if not cohesive, interaction is deleted when distance is greater than zero.
“prestress” of this link (used to simulate isotropic stress)
Up to now maximum normal strain (semi-norm), non-decreasing in time.
Normal stiffness
Shear stiffness
the damage evolution function will always return virgin state
Normal force after previous step (in global coordinates).
Damage internal variable
exponent in the rate-dependent viscoplasticity
characteristic time for viscoplasticity (if non-positive, no rate-dependence for shear)
initial length of interaction [m]
initial penetration depth of interaction [m] (used with ScGeom)
Relative ductility of bonds in normal direction
Relative residual strength
TODO
TODO
Shear force after previous step (in global coordinates).
Current normal stress
Current shear stress
tangens of internal friction angle [-]
virgin material cohesion [Pa]
Update object attributes from given dictionary
The simple linear elastic-plastic interaction with friction angle, like in the traditional [CundallStrack1979]
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Normal stiffness
Shear stiffness
Normal force after previous step (in global coordinates).
Shear force after previous step (in global coordinates).
tan of angle of friction
Update object attributes from given dictionary
is cohesion active? will be set false when a fragile contact is broken
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
true if compression plastic fracture achieved
Damage coefficient on bending, computed from maximum bending moment reached and pure creep behaviour. Its values will vary between InelastCohFrictPhys::kr and InelastCohFrictPhys::kRCrp .
Bending creep stiffness
Bending plastic unload stiffness
Tension/compression creep stiffness
Tension/compression plastic unload stiffness
Twist creep stiffness
Twist plastic unload stiffness
Normal stiffness
compression stiffness
tension stiffness
bending stiffness
shear stiffness
twist shear stiffness
Plastic failure bending moment.
Plastic failure contraction (shrinkage).
maximal bending moment reached on plastic deformation.
maximal compression reached on plastic deformation. maxCrpRchdC[0] stores un and maxCrpRchdC[1] stores Fn.
maximal extension reached on plastic deformation. maxCrpRchdT[0] stores un and maxCrpRchdT[1] stores Fn.
maximal twist reached on plastic deformation. maxCrpRchdTw[0] stores twist angle and maxCrpRchdTw[1] stores twist moment.
Maximum bending elastic moment.
Maximum compression elastic force.
Maximum tension elastic force.
Maximum twist elastic moment.
Plastic failure extension (stretching).
Plastic failure twist angle
Bending moment
Twist moment
Normal force after previous step (in global coordinates).
true if plasticity achieved on bending
true if plasticity achieved on compression
true if plasticity achieved on traction
true if plasticity achieved on twisting
Pure creep curve, used for comparison in calculation.
Maximum elastic shear force (cohesion).
Shear force after previous step (in global coordinates).
tan of angle of friction
plastic twist penetration depth describing the equilibrium state.
plastic normal penetration depth describing the equilibrium state.
Update object attributes from given dictionary
Representation of a single interaction of the JCFpm type, storage for relevant parameters
computed from tensile strength (or joint variant) to define the maximum admissible normal force in traction. [N]
computed from cohesion (or jointCohesion) to define the maximum admissible tangential force in shear, for Fn=0. [N]
crossSection=pi*Rmin^2. [m2]
Return dictionary of attributes.
defines the normal displacement in the joint after sliding treshold. [m]
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
equilibrium distance for interacting particles. Computed as the interparticular distance at first contact detection.
If false, particles interact in a frictional way. If true, particles are bonded regarding the given cohesion and tensile strength (or their jointed variants).
defined as true when both interacting particles are on joint and are in opposite sides of the joint surface. In this case, mechanical parameters of the interaction are derived from the ‘’joint...’’ material properties of the particles, and normal of the interaction is re-oriented (see also Law2_ScGeom_JCFpmPhys_JointedCohesiveFrictionalPM).
sliding distance for particles interacting on a joint. Used, when is true, to take into account dilatancy due to shearing. [-]
Normal stiffness
Shear stiffness
specifies if the interaction is crossed by more than 3 joints. If true, interaction is deleted (temporary solution).
Normal force after previous step (in global coordinates).
Shear force after previous step (in global coordinates).
tangent of the angle defining the dilatancy of the joint surface (auto. computed from JCFpmMat.jointDilationAngle). [-]
tangent of Coulomb friction angle for this interaction (auto. computed). [-]
Update object attributes from given dictionary
IPhys created from LudingMat, for use with Law2_ScGeom_LudingPhys_Basic.
Maximum overlap between particles for a collision
MinimalDelta value of delta
Force free overlap, plastic contact deformation
Maximum overlap between particles for the limit case
Max force free overlap, plastic contact deformation
Previous value of delta
Viscous damping
Dimensionless plasticity depth
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Slope of loading plastic branch
Slope of unloading and reloading elastic branch
Slope of irreversible, tensile adhesive branch
Normal stiffness
Slope of unloading and reloading limit elastic branch
Shear stiffness
Normal force after previous step (in global coordinates).
Shear force after previous step (in global coordinates).
tan of angle of friction
Update object attributes from given dictionary
Adds capillary physics to Mindlin’s interaction physics.
Defines the surface area wetted by the meniscus on the smallest grains of radius R1 (R1<R2)
Defines the surface area wetted by the meniscus on the biggest grains of radius R2 (R1<R2)
Shear force in local axes (computed incrementally)
Force of adhesion as predicted by DMT
Constant coefficient to define contact viscous damping for non-linear elastic force-displacement relationship.
Fraction of the viscous damping coefficient (normal direction) equal to .
Fraction of the viscous damping coefficient (shear direction) equal to .
Value of the capillary pressure Uc defines as Ugas-Uliquid
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Capillary Force produces by the presence of the meniscus
Indicates the number of meniscii that overlap with this one
bool to identify if the contact is adhesive, that is to say if the contact force is attractive
If true, capillary force is zero and liquid bridge is inactive.
check if the contact is sliding (useful to calculate the ratio of sliding contacts)
Normal stiffness
Constant value in the formulation of the normal stiffness
Rotational stiffness
Shear stiffness
Constant value in the formulation of the tangential stiffness
Rotational stiffness
Coefficient to determine the maximum plastic moment to apply at the contact
Presence of a meniscus if true
Artificial bending moment to provide rolling resistance in order to account for some degree of interlocking between particles
Artificial twisting moment (no plastic condition can be applied at the moment)
Normal force after previous step (in global coordinates).
Normal viscous component
Previous local displacement; only used with Law2_L3Geom_FrictPhys_HertzMindlin.
Contact radius (only computed with Law2_ScGeom_MindlinPhys_Mindlin::calcEnergy)
Total elastic shear force
Shear force after previous step (in global coordinates).
Shear viscous component
tan of angle of friction
Update object attributes from given dictionary
Total elastic shear displacement (only elastic part)
Total elastic shear displacement (elastic+plastic part)
Volume of the menicus
Representation of an interaction of the Hertz-Mindlin type.
Shear force in local axes (computed incrementally)
Force of adhesion as predicted by DMT
Constant coefficient to define contact viscous damping for non-linear elastic force-displacement relationship.
Fraction of the viscous damping coefficient (normal direction) equal to .
Fraction of the viscous damping coefficient (shear direction) equal to .
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
bool to identify if the contact is adhesive, that is to say if the contact force is attractive
check if the contact is sliding (useful to calculate the ratio of sliding contacts)
Normal stiffness
Constant value in the formulation of the normal stiffness
Rotational stiffness
Shear stiffness
Constant value in the formulation of the tangential stiffness
Rotational stiffness
Coefficient to determine the maximum plastic moment to apply at the contact
Artificial bending moment to provide rolling resistance in order to account for some degree of interlocking between particles
Artificial twisting moment (no plastic condition can be applied at the moment)
Normal force after previous step (in global coordinates).
Normal viscous component
Previous local displacement; only used with Law2_L3Geom_FrictPhys_HertzMindlin.
Contact radius (only computed with Law2_ScGeom_MindlinPhys_Mindlin::calcEnergy)
Total elastic shear force
Shear force after previous step (in global coordinates).
Shear viscous component
tan of angle of friction
Update object attributes from given dictionary
Total elastic shear displacement (only elastic part)
Total elastic shear displacement (elastic+plastic part)
Abstract class for interactions that have normal stiffness.
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Normal stiffness
Normal force after previous step (in global coordinates).
Update object attributes from given dictionary
Abstract class for interactions that have shear stiffnesses, in addition to normal stiffness. This class is used in the PFC3d-style stiffness timestepper.
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Normal stiffness
Shear stiffness
Normal force after previous step (in global coordinates).
Shear force after previous step (in global coordinates).
Update object attributes from given dictionary
Physics (of interaction) for using Law2_ScGeom6D_NormalInelasticityPhys_NormalInelasticity : with inelastic unloadings
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
parameter stored for each interaction, and allowing to compute the maximum value of the exchanged torque : TorqueMax= forMaxMoment * NormalForce
Normal stiffness
the stifness corresponding to a virgin load for example
the rolling stiffness of the interaction
Shear stiffness
Bending moment. Defined here, being initialized as it should be, to be used in Law2_ScGeom6D_NormalInelasticityPhys_NormalInelasticity
Twist moment. Defined here, being initialized as it should be, to be used in Law2_ScGeom6D_NormalInelasticityPhys_NormalInelasticity
Normal force after previous step (in global coordinates).
the value of the normal force at the last time step
the value of this un at the last time step
Shear force after previous step (in global coordinates).
tan of angle of friction
the maximum value of penetration depth of the history of this interaction
Update object attributes from given dictionary
Simple elastic material with friction for volumetric constitutive laws
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Normal stiffness
Shear stiffness
Normal force after previous step (in global coordinates).
Shear force after previous step (in global coordinates).
tangens of angle of internal friction
Update object attributes from given dictionary
IPhys created from ViscElMat, for use with Law2_ScGeom_ViscElPhys_Basic.
True, if capillar forces need to be added.
Different types of capillar interaction: Willett_numeric, Willett_analytic, Weigert, Rabinovich
Liquid bridge volume [m^3]
Normal viscous constant
Shear viscous constant
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Surface tension [N/m]
Normal stiffness
Shear stiffness
Whether liquid bridge was created, only after a normal contact of spheres
Rolling resistance, see [Zhou1999536].
Rolling resistance type, see [Zhou1999536]. mRtype=1 - equation (3) in [Zhou1999536]; mRtype=2 - equation (4) in [Zhou1999536]
Normal force after previous step (in global coordinates).
Critical bridge length [m]
Shear force after previous step (in global coordinates).
tan of angle of friction
Contact angle [rad]
Update object attributes from given dictionary
Temporary version of FrictPhys for compatibility with e.g. Law2_ScGeom6D_NormalInelasticityPhys_NormalInelasticity
Creeped force (parallel)
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Normal stiffness
Shear stiffness
Normal force after previous step (in global coordinates).
Shear force after previous step (in global coordinates).
tan of angle of friction
Update object attributes from given dictionary
Representation of a single interaction of the WirePM type, storage for relevant parameters
Additional wire length for considering the distortion for WireMat type=2 (see [Thoeni2013]).
Return dictionary of attributes.
Return list of dispatch classes (from down upwards), starting with the class instance itself, top-level indexable at last. If names is true (default), return class names rather than numerical indices.
Return class index of this instance.
Defines the values for force-displacement curve.
Equilibrium distance for particles. Computed as the initial inter-particular distance when particle are linked.
If true the properties of the interaction will be defined as a double-twisted wire.
If true particles are linked and will interact. Interactions are linked automatically by the definition of the corresponding interaction radius. The value is false if the wire breaks (no more interaction).
Normal stiffness
Shear stiffness
This value indicates on how far from failing the wire is, e.g. actual normal displacement divided by admissible normal displacement.
Normal force after previous step (in global coordinates).
Plastic part of the inter-particular distance of the previous step.
Note
Only elastic displacements are reversible (the elastic stiffness is used for unloading) and compressive forces are inadmissible. The compressive stiffness is assumed to be equal to zero.
Shear force after previous step (in global coordinates).
Defines the values for the various stiffnesses (the elastic stiffness is stored as kn).
tan of angle of friction
Update object attributes from given dictionary
Engine that will generally affect the whole simulation (contrary to PartialEngine).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Box geometry of the SpheresFactory region, given by extents and center
PSD-Input is in mass (true), otherwise the number of particles will be considered.
PSD-dispersion, cumulative procent meanings [-]
PSD-dispersion, sizes of cells, Diameter [m]
Blocked degress of freedom
Center of the region
Use the color for newly created particles, if specified
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
If true, the particles only with the defined in PSDsizes diameters will be created. Otherwise the diameter will be randomly chosen in the range [PSDsizes[i-1]:PSDsizes[i]], in this case the length of PSDsizes should be more on 1, than the length of PSDcum.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Extents of the region
Total mass that should be attained at the end of the current step. (auto-updated)
ids of created bodies
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
groupMask to apply for newly created spheres
Mass flow rate [kg/s]
Shared material id to use for newly created spheres (can be negative to count from the end)
Maximum number of attempts to position a new sphere randomly.
Maximal mass at which to stop generating new particles regardless of massFlowRate. if maxMass=-1 - this parameter is ignored.
The number of particles at which to stop generating new ones regardless of massFlowRate. if maxParticles=-1 - this parameter is ignored .
Orientation of the region’s geometry, direction of particle’s velocites if normalVel is not set.
Direction of particle’s velocites.
Cummulative number of particles produces so far (auto-updated)
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Maximum radius of generated spheres (uniform distribution)
Minimum radius of generated spheres (uniform distribution)
If true no complain about excessing maxAttempt but disable the factory (by set massFlowRate=0).
If true, the SpheresFactory stops (sets massFlowRate=0), when maximal number of attempts to insert particle exceed.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Mass of spheres that was produced so far. (auto-updated)
Volume of spheres that was produced so far. (auto-updated)
Update object attributes from given dictionary
Maximum angle by which the initial sphere velocity deviates from the normal.
Maximum velocity norm of generated spheres (uniform distribution)
Minimum velocity norm of generated spheres (uniform distribution)
Records information from capillary meniscii on samples submitted to triaxial compressions. -> New formalism needs to be tested!!!
Adds an iteration number to the file name, when the file was created. Useful for creating new files at each call (false by default)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Name of file to save to; must not be empty.
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Whether to delete current file contents, if any, when opening (false by default)
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Circular geometry of the SpheresFactory region. It can be disk (given by radius and center), or cylinder (given by radius, length and center).
PSD-Input is in mass (true), otherwise the number of particles will be considered.
PSD-dispersion, cumulative procent meanings [-]
PSD-dispersion, sizes of cells, Diameter [m]
Blocked degress of freedom
Center of the region
Use the color for newly created particles, if specified
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
If true, the particles only with the defined in PSDsizes diameters will be created. Otherwise the diameter will be randomly chosen in the range [PSDsizes[i-1]:PSDsizes[i]], in this case the length of PSDsizes should be more on 1, than the length of PSDcum.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Total mass that should be attained at the end of the current step. (auto-updated)
ids of created bodies
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Length of the cylindrical region (0 by default)
groupMask to apply for newly created spheres
Mass flow rate [kg/s]
Shared material id to use for newly created spheres (can be negative to count from the end)
Maximum number of attempts to position a new sphere randomly.
Maximal mass at which to stop generating new particles regardless of massFlowRate. if maxMass=-1 - this parameter is ignored.
The number of particles at which to stop generating new ones regardless of massFlowRate. if maxParticles=-1 - this parameter is ignored .
Orientation of the region’s geometry, direction of particle’s velocites if normalVel is not set.
Direction of particle’s velocites.
Cummulative number of particles produces so far (auto-updated)
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Maximum radius of generated spheres (uniform distribution)
Minimum radius of generated spheres (uniform distribution)
Radius of the region
If true no complain about excessing maxAttempt but disable the factory (by set massFlowRate=0).
If true, the SpheresFactory stops (sets massFlowRate=0), when maximal number of attempts to insert particle exceed.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Mass of spheres that was produced so far. (auto-updated)
Volume of spheres that was produced so far. (auto-updated)
Update object attributes from given dictionary
Maximum angle by which the initial sphere velocity deviates from the normal.
Maximum velocity norm of generated spheres (uniform distribution)
Minimum velocity norm of generated spheres (uniform distribution)
[DEPRECATED] Loop over interactions applying Law2_ScGeom6D_CohFrictPhys_CohesionMoment on all interactions.
Note
Use InteractionLoop and Law2_ScGeom6D_CohFrictPhys_CohesionMoment instead of this class for performance reasons.
If true, use bending/twisting moments at all contacts. If false, compute moments only for cohesive contacts.
creep viscosity [Pa.s/m]. probably should be moved to Ip2_CohFrictMat_CohFrictMat_CohFrictPhys...
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
activate creep on the shear force, using CohesiveFrictionalContactLaw::creep_viscosity.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
activate creep on the twisting moment, using CohesiveFrictionalContactLaw::creep_viscosity.
Update object attributes from given dictionary
Update CpmState of bodies based on state variables in CpmPhys of interactions with this bod. In particular, bodies’ colors and CpmState::normDmg depending on average damage of their interactions and number of interactions that were already fully broken and have disappeared is updated. This engine contains its own loop (2 loops, more precisely) over all bodies and should be run periodically to update colors during the simulation, if desired.
Average residual strength at last run.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Globally maximum damage parameter at last run.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Delete particles that are out of axis-aligned box given by lo and hi.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Upper corner of the domain.
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Lower corner of the domain.
Mass of deleted particles.
If mask is defined, only particles with corresponding groupMask will be deleted.
Cummulative number of particles deleted.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Volume of deleted particles.
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
[DEPRECATED] Loop over interactions applying Law2_ScGeom_FrictPhys_CundallStrack on all interactions.
Note
Use InteractionLoop and Law2_ScGeom_FrictPhys_CundallStrack instead of this class for performance reasons.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Initializer for filling adjacency geometry data for facets.
Common vertices and common edges are identified and mutual angle between facet faces is written to Facet instances. If facets don’t move with respect to each other, this must be done only at the beginng.
how many common edges were identified during last run. (auto-updated)
how many common vertices were identified during last run. (auto-updated)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Axis along which to do the initial vertex sort
maximum distance of ‘identical’ vertices, relative to minimum facet size
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine saves the resultant force affecting to bodies, listed in ids. For instance, can be useful for defining the forces, which affects to _buldozer_ during its work.
Adds an iteration number to the file name, when the file was created. Useful for creating new files at each call (false by default)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Name of file to save to; must not be empty.
List of bodies whose state will be measured
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Resultant force, returning by the function.
Whether to delete current file contents, if any, when opening (false by default)
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Reset all forces stored in Scene::forces (O.forces in python). Typically, this is the first engine to be run at every step. In addition, reset those energies that should be reset, if energy tracing is enabled.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
An engine assigning the time-step as a fraction of the minimum eigen-period in the problem. The derivation is detailed in the chapter on DEM formulation. The viscEl option enables to evaluate the timestep in a similar way for the visco-elastic contact law Law2_ScGeom_ViscElPhys_Basic, more detail in GlobalStiffnessTimestepper::viscEl.
is the engine active?
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
used as the initial value of the timestep (especially useful in the first steps when no contact exist). If negative, it will be defined by utils.PWaveTimeStep * GlobalStiffnessTimeStepper::timestepSafetyCoefficient
(auto-updated) don’t modify this value if you don’t plan to modify the scaling factor manually for some bodies. In most cases, it is enough to set NewtonIntegrator::densityScaling and let this one be adjusted automatically.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
if positive, used as max value of the timestep whatever the computed value
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
last computed dt (auto-updated)
if NewtonIntegrator::densityScaling is active, this value will be used as the simulation timestep and the scaling will use this value of dt as the target value. The value of targetDt is arbitrary and should have no effect in the result in general. However if some bodies have imposed velocities, for instance, they will move more or less per each step depending on this value.
dt update interval
safety factor between the minimum eigen-period and the final assigned dt (less than 1))
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
To use with ViscElPhys. if True, evaluate separetly the minimum eigen-period in the problem considering only the elastic contribution on one hand (spring only), and only the viscous contribution on the other hand (dashpot only). Take then the minimum of the two and use the safety coefficient GlobalStiffnessTimestepper::timestepSafetyCoefficient to take into account the possible coupling between the two contribution.
Unified dispatcher for handling interaction loop at every step, for parallel performance reasons.
Special constructor
Constructs from 3 lists of Ig2, Ip2, Law functors respectively; they will be passed to interal dispatchers, which you might retrieve.
Callbacks which will be called for every Interaction, if activated.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Defines if the interaction loop should erase pending interactions, else the collider takes care of that alone (depends on what collider is used).
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
IGeomDispatcher object that is used for dispatch.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
LawDispatcher object used for dispatch.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
IPhysDispatcher object used for dispatch.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
This law allows one to take into account capillary forces/effects between spheres coming from the presence of interparticular liquid bridges (menisci).
The control parameter is the capillary pressure (or suction) Uc = ugas - Uliquid. Liquid bridges properties (volume V, extent over interacting grains delta1 and delta2) are computed as a result of the defined capillary pressure and of the interacting geometry (spheres radii and interparticular distance).
References: in english [Scholtes2009b]; more detailed, but in french [Scholtes2009d].
The law needs ascii files M(r=i) with i=R1/R2 to work (see https://yade-dem.org/index.php/CapillaryTriaxialTest). These ASCII files contain a set of results from the resolution of the Laplace-Young equation for different configurations of the interacting geometry.
In order to allow capillary forces between distant spheres, it is necessary to enlarge the bounding boxes using Bo1_Sphere_Aabb::aabbEnlargeFactor and make the Ig2 define define distant interactions via interactionDetectionFactor. It is also necessary to disable interactions removal by the constitutive law (Law2). The only combinations of laws supported are currently capillary law + Law2_ScGeom_FrictPhys_CundallStrack and capillary law + Law2_ScGeom_MindlinPhys_Mindlin (and the other variants of Hertz-Mindlin).
See CapillaryPhys-example.py for an example script.
If true, capillary forces are set to zero as soon as, at least, 1 overlap (menisci fusion) is detected
Value of the capillary pressure Uc defines as Uc=Ugas-Uliquid
Generate meniscii between distant spheres? Else only maintain the existing ones. For modeling a wetting path this flag should always be false. For a drying path it should be true for one step (initialization) then false, as in the logic of [Scholtes2009c]
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
If true potential menisci overlaps are checked
(auto-updated) true if hertz model is used
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Compute fabric tensor, local porosity, local deformation, and other micromechanicaly defined quantities based on triangulation/tesselation of the packing.
Is the engine just saving states or also computing and outputing deformations for each increment?
Should increments of force and displacements be defined on [n,n+1]? If not, states will be saved with only positions and forces (no displacements).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Number of timesteps between analyzed states.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
bodies that are not spheres will be used to defines bounds (else just skipped).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Base name for increment analysis output file.
Base name of state files.
A number incremented and appended at the end of output files to reflect increment number.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine integrating newtonian motion equations.
damping coefficient for Cundall’s non viscous damping (see [Chareyre2005]) [-]
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
if True, then density scaling [Pfc3dManual30] will be applied in order to have a critical timestep equal to GlobalStiffnessTimeStepper::targetDt for all bodies. This option makes the simulation unrealistic from a dynamic point of view, but may speedup quasistatic simulations. In rare situations, it could be useful to not set the scalling factor automatically for each body (which the time-stepper does). In such case revert GlobalStiffnessTimeStepper.densityScaling to False.
Return dictionary of attributes.
Enable more exact body rotation integrator for aspherical bodies only, using formulation from [Allen1989], pg. 89.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Gravitational acceleration (effectively replaces GravityEngine).
Whether to separately track translational and rotational kinetic energy.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
store square of max. velocity, for informative purposes; computed again at every step. (auto-updated)
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Store previous velocity gradient (Cell::velGrad) to track acceleration. (auto-updated)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Warn when forces were not resetted in this step by ForceResetter; this mostly points to ForceResetter being forgotten incidentally and should be disabled only with a good reason.
Run Engine::action with given fixed periodicity real time (=wall clock time, computation time), virtual time (simulation time), iteration number), by setting any of those criteria (virtPeriod, realPeriod, iterPeriod) to a positive value. They are all negative (inactive) by default.
The number of times this engine is activated can be limited by setting nDo>0. If the number of activations will have been already reached, no action will be called even if an active period has elapsed.
If initRun is set (false by default), the engine will run when called for the first time; otherwise it will only start counting period (realLast etc interal variables) from that point, but without actually running, and will run only once a period has elapsed since the initial run.
This class should not be used directly; rather, derive your own engine which you want to be run periodically.
Derived engines should override Engine::action(), which will be called periodically. If the derived Engine overrides also Engine::isActivated, it should also take in account return value from PeriodicEngine::isActivated, since otherwise the periodicity will not be functional.
Example with PyRunner, which derives from PeriodicEngine; likely to be encountered in python scripts:
PyRunner(realPeriod=5,iterPeriod=10000,command='print O.iter')
will print iteration number every 10000 iterations or every 5 seconds of wall clock time, whiever comes first since it was last run.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Engine that splits polyhedras.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Execute a python command periodically, with defined (and adjustable) periodicity. See PeriodicEngine documentation for details.
Command to be run by python interpreter. Not run if empty.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Engine periodically storing some data to (one) external file. In addition PeriodicEngine, it handles opening the file as needed. See PeriodicEngine for controlling periodicity.
Adds an iteration number to the file name, when the file was created. Useful for creating new files at each call (false by default)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Name of file to save to; must not be empty.
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Whether to delete current file contents, if any, when opening (false by default)
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Creates spheres during simulation, placing them at random positions. Every time called, one new sphere will be created and inserted in the simulation.
Mean angularVelocity of spheres.
Half size of a angularVelocity distribution interval. New sphere will have random angularVelocity within the range angularVelocity±angularVelocityRange.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
The geometry of the section where spheres will be placed; they will be placed on facets or in volume between them depending on volumeSection flag.
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Max attempts to place sphere. If placing the sphere in certain random position would cause an overlap with any other physical body in the model, SpheresFactory will try to find another position.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
??
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
??
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Affected bodies.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Mean velocity of spheres.
Half size of a velocities distribution interval. New sphere will have random velocity within the range velocity±velocityRange.
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Create new spheres inside factory volume rather than on its surface.
Periodically save snapshots of GLView(s) as .png files. Files are named fileBase + counter + ‘.png’ (counter is left-padded by 0s, i.e. snap00004.png).
Number that will be appended to fileBase when the next snapshot is saved (incremented at every save). (auto-updated)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Timeout for 3d operations (opening new view, saving snapshot); after timing out, throw exception (or only report error if ignoreErrors) and make myself dead. [s]
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Basename for snapshots
Format of snapshots (one of JPEG, PNG, EPS, PS, PPM, BMP) QGLViewer documentation. File extension will be lowercased format. Validity of format is not checked.
Only report errors instead of throwing exceptions, in case of timeouts.
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
number of msec to sleep after snapshot (to prevent 3d hw problems) [ms]
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Name of field in plot.imgData to which taken snapshots will be appended automatically.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Files that have been created so far
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Engine for spitting spheres based on mass flow rate, particle size distribution etc. Initial velocity of particles is given by vMin, vMax, the massFlowRate determines how many particles to generate at each step. When goalMass is attained or positive maxParticles is reached, the engine does not produce particles anymore. Geometry of the region should be defined in a derived engine by overridden SpheresFactory::pickRandomPosition().
A sample script for this engine is in scripts/spheresFactory.py.
PSD-Input is in mass (true), otherwise the number of particles will be considered.
PSD-dispersion, cumulative procent meanings [-]
PSD-dispersion, sizes of cells, Diameter [m]
Blocked degress of freedom
Use the color for newly created particles, if specified
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
If true, the particles only with the defined in PSDsizes diameters will be created. Otherwise the diameter will be randomly chosen in the range [PSDsizes[i-1]:PSDsizes[i]], in this case the length of PSDsizes should be more on 1, than the length of PSDcum.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Total mass that should be attained at the end of the current step. (auto-updated)
ids of created bodies
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
groupMask to apply for newly created spheres
Mass flow rate [kg/s]
Shared material id to use for newly created spheres (can be negative to count from the end)
Maximum number of attempts to position a new sphere randomly.
Maximal mass at which to stop generating new particles regardless of massFlowRate. if maxMass=-1 - this parameter is ignored.
The number of particles at which to stop generating new ones regardless of massFlowRate. if maxParticles=-1 - this parameter is ignored .
Orientation of the region’s geometry, direction of particle’s velocites if normalVel is not set.
Direction of particle’s velocites.
Cummulative number of particles produces so far (auto-updated)
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Maximum radius of generated spheres (uniform distribution)
Minimum radius of generated spheres (uniform distribution)
If true no complain about excessing maxAttempt but disable the factory (by set massFlowRate=0).
If true, the SpheresFactory stops (sets massFlowRate=0), when maximal number of attempts to insert particle exceed.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Mass of spheres that was produced so far. (auto-updated)
Volume of spheres that was produced so far. (auto-updated)
Update object attributes from given dictionary
Maximum angle by which the initial sphere velocity deviates from the normal.
Maximum velocity norm of generated spheres (uniform distribution)
Minimum velocity norm of generated spheres (uniform distribution)
Handle the triangulation of spheres in a scene, build tesselation on request, and give access to computed quantities (see also the dedicated section in user manual). The calculation of microstrain is explained in [Catalano2013a]
See example usage in script example/tesselationWrapper/tesselationWrapper.py.
Below is an output of the defToVtk function visualized with paraview (in this case Yade’s TesselationWrapper was used to process experimental data obtained on sand by Edward Ando at Grenoble University, 3SR lab.)
Compute per-particle deformation. Get it with TesselationWrapper::deformation (id,i,j).
Compute volumes of all Voronoi’s cells.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Write local deformations in vtk format from states 0 and 1.
Write local deformations in vtk format from positions files (one sphere per line, with x,y,z,rad separated by spaces).
Write local deformations in vtk format from state files (since the file format is very special, consider using defToVtkFromPositions if the input files were not generated by TesselationWrapper).
Get particle deformation
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Return a table with per-sphere computed quantities. Include deformations on the increment defined by states 0 and 1 if deformation=True (make sure to define states 0 and 1 consistently).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Load a file with positions to define state 0 or 1.
(auto-computed)
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Save a file with positions, can be later reloaded in order to define state 0 or 1.
Make the current state of the simulation the initial (0) or final (1) configuration for the definition of displacement increments, use only state=0 if you just want to get volmumes and porosity.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
triangulate spheres of the packing
Update object attributes from given dictionary
Returns the volume of Voronoi’s cell of a sphere.
Calculate physical response of 2 tetrahedra in interaction, based on penetration configuration given by TTetraGeom.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine defining time-step (fundamental class)
is the engine active?
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
dt update interval
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine saves the total torque according to the given axis and ZeroPoint, the force is taken from bodies, listed in ids For instance, can be useful for defining the torque, which affects on ball mill during its work.
Adds an iteration number to the file name, when the file was created. Useful for creating new files at each call (false by default)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Name of file to save to; must not be empty.
List of bodies whose state will be measured
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Rotation axis
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Resultant torque, returning by the function.
Whether to delete current file contents, if any, when opening (false by default)
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Point of rotation center
Engine recording triaxial variables (see the variables list in the first line of the output file). This recorder needs TriaxialCompressionEngine or ThreeDTriaxialEngine present in the simulation).
Adds an iteration number to the file name, when the file was created. Useful for creating new files at each call (false by default)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Name of file to save to; must not be empty.
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
porosity of the packing [-]
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Whether to delete current file contents, if any, when opening (false by default)
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Engine recording snapshots of simulation into series of *.vtu files, readable by VTK-based postprocessing programs such as Paraview. Both bodies (spheres and facets) and interactions can be recorded, with various vector/scalar quantities that are defined on them.
PeriodicEngine.initRun is initialized to True automatically.
Necessary if recorders contains ‘cracks’. A string specifying the name of file ‘cracks___.txt’ that is considered in this case (see corresponding attribute).
Store data as readable text in the XML file (sets vtkXMLWriter data mode to vtkXMLWriter::Ascii, while the default is Appended
Compress output XML files [experimental].
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Base file name; it will be appended with {spheres,intrs,facets}-243100.vtu (unless multiblock is True) depending on active recorders and step number (243100 in this case). It can contain slashes, but the directory must exist already.
Run the first time we are called as well.
Tracks step number of last run (auto-updated).
Periodicity criterion using step number (deactivated if <= 0)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
If mask defined, only bodies with corresponding groupMask will be exported. If 0, all bodies will be exported.
Use multi-block (.vtm) files to store data, rather than separate .vtu files.
Limit number of executions by this number (deactivated if negative)
Track number of executions (cummulative) (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Tracks real time of last run (auto-updated).
Periodicity criterion using real (wall clock, computation, human) time (deactivated if <=0)
List of active recorders (as strings). all (the default value) enables all base and generic recorders.
Base recorders
Base recorders save the geometry (unstructured grids) on which other data is defined. They are implicitly activated by many of the other recorders. Each of them creates a new file (or a block, if multiblock is set).
Generic recorders
Generic recorders do not depend on specific model being used and save commonly useful data.
Specific recorders
The following should only be activated in appropriate cases, otherwise crashes can occur due to violation of type presuppositions.
- cpm
- Saves data pertaining to the concrete model: cpmDamage (normalized residual strength averaged on particle), cpmStress (stress on particle); intr is activated automatically by cpm
- wpm
- Saves data pertaining to the wire particle model: wpmForceNFactor shows the loading factor for the wire, e.g. normal force divided by threshold normal force.
- jcfpm
- Saves data pertaining to the rock (smooth)-jointed model: damage is defined by JCFpmState.tensBreak + JCFpmState.shearBreak; intr is activated automatically by jcfpm, and on joint or cohesive interactions can be vizualized.
- cracks
- Saves other data pertaining to the rock model: cracks shows locations where cohesive bonds failed during the simulation, with their types (0/1 for tensile/shear breakages), their sizes (0.5*(R1+R2)), and their normal directions. The corresponding attribute has to be activated, and Key attributes have to be consistent.
Skip interactions with facets, when saving interactions
Skip non-dynamic spheres (but not facets).
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Tracks virtual time of last run (auto-updated).
Periodicity criterion using virtual (simulation) time (deactivated if <= 0)
Base for engines controlling boundary conditions of simulations. Not to be used directly.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Disturbs a simple shear sample in a given displacement direction
This engine allows one to apply, on a simple shear sample, a loading controlled by du/dgamma = cste, which is equivalent to du + cste’ * dgamma = 0 (proportionnal path loadings). To do so, the upper plate of the simple shear box is moved in a given direction (corresponding to a given du/dgamma), whereas lateral plates are moved so that the box remains closed. This engine can easily be used to perform directionnal probes, with a python script launching successivly the same .xml which contains this engine, after having modified the direction of loading (see theta attribute). That’s why this Engine contains a saveData procedure which can save data on the state of the sample at the end of the loading (in case of successive loadings - for successive directions - through a python script, each line would correspond to one direction of loading).
string to add at the names of the saved files, and of the output file filled by saveData
boolean controling the output of messages on the screen
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
the id of the wall at the back of the sample
the id of the lower wall
the id of the wall in front of the sample
the id of the left wall
the id of the right wall
the id of the upper wall
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
the number of iterations of loading to perform
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
the angle, in a (gamma,h=-u) plane from the gamma - axis to the perturbation vector (trigo wise) [degrees]
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
the speed at which the perturbation is imposed. In case of samples which are more sensitive to normal loadings than tangential ones, one possibility is to take v = V_shear - | (V_shear-V_comp)*sin(theta) | => v=V_shear in shear; V_comp in compression [m/s]
To apply a Constant Normal Displacement (CND) shear for a parallelogram box
This engine, designed for simulations implying a simple shear box (SimpleShear Preprocessor or scripts/simpleShear.py), allows one to perform a constant normal displacement shear, by translating horizontally the upper plate, while the lateral ones rotate so that they always keep contact with the lower and upper walls.
string to add at the names of the saved files
boolean controling the output of messages on the screen
the angle from the lower box to the left box (trigo wise). Measured by this Engine. Has to be saved, but not to be changed by the user.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
the (vertical) force acting on the upper plate on the very first time step (determined by the Engine). Controls of the loadings in case of KinemCNSEngine or KinemCNLEngine will be done according to this initial value []. Has to be saved, but not to be changed by the user.
boolean set to false as soon as the engine has done its job one time : useful to know if initial height of, and normal force sustained by, the upper box are known or not (and thus if they have to be initialized). Has to be saved, but not to be changed by the user.
the current value of the tangential displacement
vector with the values of gamma at which a save of the simulation is performed [m]
the value of the tangential displacement at wich the displacement is stopped [m]
the id of the wall at the back of the sample
the id of the lower wall
the id of the wall in front of the sample
the id of the left wall
the id of the right wall
the id of the upper wall
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
to limit the speed of the vertical displacements done to control (CNL or CNS cases) [
]
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
the speed at which the shear is performed : speed of the upper plate [m/s]
vector (same length as ‘gamma_save’ for ex), with 0 or 1 depending whether the save for the corresponding value of gamma has been done (1) or not (0). Has to be saved, but not to be changed by the user.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
the vertical displacements done to to control (CNL or CNS cases) are in fact damped, through this wallDamping
the height of the upper plate at the very first time step : the engine finds its value []. Has to be saved, but not to be changed by the user.
To apply a constant normal stress shear (i.e. Constant Normal Load : CNL) for a parallelogram box (simple shear box : SimpleShear Preprocessor or scripts/simpleShear.py)
This engine allows one to translate horizontally the upper plate while the lateral ones rotate so that they always keep contact with the lower and upper walls.
In fact the upper plate can move not only horizontally but also vertically, so that the normal stress acting on it remains constant (this constant value is not chosen by the user but is the one that exists at the beginning of the simulation)
The right vertical displacements which will be allowed are computed from the rigidity Kn of the sample over the wall (so to cancel a deltaSigma, a normal dplt deltaSigma*S/(Kn) is set)
The movement is moreover controlled by the user via a shearSpeed which will be the speed of the upper wall, and by a maximum value of horizontal displacement gammalim, after which the shear stops.
Note
Not only the positions of walls are updated but also their speeds, which is all but useless considering the fact that in the contact laws these velocities of bodies are used to compute values of tangential relative displacements.
Warning
Because of this last point, if you want to use later saves of simulations executed with this Engine, but without that stopMovement was executed, your boxes will keep their speeds => you will have to cancel them ‘by hand’ in the .xml.
string to add at the names of the saved files
boolean controling the output of messages on the screen
the angle from the lower box to the left box (trigo wise). Measured by this Engine. Has to be saved, but not to be changed by the user.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
the (vertical) force acting on the upper plate on the very first time step (determined by the Engine). Controls of the loadings in case of KinemCNSEngine or KinemCNLEngine will be done according to this initial value []. Has to be saved, but not to be changed by the user.
boolean set to false as soon as the engine has done its job one time : useful to know if initial height of, and normal force sustained by, the upper box are known or not (and thus if they have to be initialized). Has to be saved, but not to be changed by the user.
current value of tangential displacement [m]
vector with the values of gamma at which a save of the simulation is performed [m]
the value of tangential displacement (of upper plate) at wich the shearing is stopped [m]
the id of the wall at the back of the sample
the id of the lower wall
the id of the wall in front of the sample
the id of the left wall
the id of the right wall
the id of the upper wall
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
to limit the speed of the vertical displacements done to control (CNL or CNS cases) [
]
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
the speed at wich the shearing is performed : speed of the upper plate [m/s]
vector (same length as ‘gamma_save’ for ex), with 0 or 1 depending whether the save for the corresponding value of gamma has been done (1) or not (0). Has to be saved, but not to be changed by the user.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
the vertical displacements done to to control (CNL or CNS cases) are in fact damped, through this wallDamping
the height of the upper plate at the very first time step : the engine finds its value []. Has to be saved, but not to be changed by the user.
To apply a Constant Normal Stifness (CNS) shear for a parallelogram box (simple shear)
This engine, useable in simulations implying one deformable parallelepipedic box, allows one to translate horizontally the upper plate while the lateral ones rotate so that they always keep contact with the lower and upper walls. The upper plate can move not only horizontally but also vertically, so that the normal rigidity defined by DeltaF(upper plate)/DeltaU(upper plate) = constant (= KnC defined by the user).
The movement is moreover controlled by the user via a shearSpeed which is the horizontal speed of the upper wall, and by a maximum value of horizontal displacement gammalim (of the upper plate), after which the shear stops.
Note
not only the positions of walls are updated but also their speeds, which is all but useless considering the fact that in the contact laws these velocities of bodies are used to compute values of tangential relative displacements.
Warning
But, because of this last point, if you want to use later saves of simulations executed with this Engine, but without that stopMovement was executed, your boxes will keep their speeds => you will have to cancel them by hand in the .xml
string to add at the names of the saved files
the normal rigidity chosen by the user [MPa/mm] - the conversion in Pa/m will be made
boolean controling the output of messages on the screen
the angle from the lower box to the left box (trigo wise). Measured by this Engine. Has to be saved, but not to be changed by the user.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
the (vertical) force acting on the upper plate on the very first time step (determined by the Engine). Controls of the loadings in case of KinemCNSEngine or KinemCNLEngine will be done according to this initial value []. Has to be saved, but not to be changed by the user.
boolean set to false as soon as the engine has done its job one time : useful to know if initial height of, and normal force sustained by, the upper box are known or not (and thus if they have to be initialized). Has to be saved, but not to be changed by the user.
current value of tangential displacement [m]
the value of tangential displacement (of upper plate) at wich the shearing is stopped [m]
the id of the wall at the back of the sample
the id of the lower wall
the id of the wall in front of the sample
the id of the left wall
the id of the right wall
the id of the upper wall
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
to limit the speed of the vertical displacements done to control (CNL or CNS cases) [
]
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
the speed at wich the shearing is performed : speed of the upper plate [m/s]
vector (same length as ‘gamma_save’ for ex), with 0 or 1 depending whether the save for the corresponding value of gamma has been done (1) or not (0). Has to be saved, but not to be changed by the user.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
the vertical displacements done to to control (CNL or CNS cases) are in fact damped, through this wallDamping
the height of the upper plate at the very first time step : the engine finds its value []. Has to be saved, but not to be changed by the user.
To compress a simple shear sample by moving the upper box in a vertical way only, so that the tangential displacement (defined by the horizontal gap between the upper and lower boxes) remains constant (thus, the CTD = Constant Tangential Displacement). The lateral boxes move also to keep always contact. All that until this box is submitted to a given stress (=*targetSigma*). Moreover saves are executed at each value of stresses stored in the vector sigma_save, and at targetSigma
string to add at the names of the saved files
boolean controling the output of messages on the screen
the angle from the lower box to the left box (trigo wise). Measured by this Engine. Has to be saved, but not to be changed by the user.
(vertical) speed of the upper box : >0 for real compression, <0 for unloading []
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
the (vertical) force acting on the upper plate on the very first time step (determined by the Engine). Controls of the loadings in case of KinemCNSEngine or KinemCNLEngine will be done according to this initial value []. Has to be saved, but not to be changed by the user.
boolean set to false as soon as the engine has done its job one time : useful to know if initial height of, and normal force sustained by, the upper box are known or not (and thus if they have to be initialized). Has to be saved, but not to be changed by the user.
the id of the wall at the back of the sample
the id of the lower wall
the id of the wall in front of the sample
the id of the left wall
the id of the right wall
the id of the upper wall
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
to limit the speed of the vertical displacements done to control (CNL or CNS cases) [
]
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
vector with the values of sigma at which a save of the simulation should be performed []
the value of sigma at which the compression should stop []
vector (same length as ‘gamma_save’ for ex), with 0 or 1 depending whether the save for the corresponding value of gamma has been done (1) or not (0). Has to be saved, but not to be changed by the user.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
the vertical displacements done to to control (CNL or CNS cases) are in fact damped, through this wallDamping
the height of the upper plate at the very first time step : the engine finds its value []. Has to be saved, but not to be changed by the user.
This class is supposed to be a mother class for all Engines performing loadings on the simple shear box of SimpleShear. It is not intended to be used by itself, but its declaration and implentation will thus contain all what is useful for all these Engines. The script simpleShear.py illustrates the use of the various corresponding Engines.
string to add at the names of the saved files
boolean controling the output of messages on the screen
the angle from the lower box to the left box (trigo wise). Measured by this Engine. Has to be saved, but not to be changed by the user.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
the (vertical) force acting on the upper plate on the very first time step (determined by the Engine). Controls of the loadings in case of KinemCNSEngine or KinemCNLEngine will be done according to this initial value []. Has to be saved, but not to be changed by the user.
boolean set to false as soon as the engine has done its job one time : useful to know if initial height of, and normal force sustained by, the upper box are known or not (and thus if they have to be initialized). Has to be saved, but not to be changed by the user.
the id of the wall at the back of the sample
the id of the lower wall
the id of the wall in front of the sample
the id of the left wall
the id of the right wall
the id of the upper wall
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
to limit the speed of the vertical displacements done to control (CNL or CNS cases) [
]
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
vector (same length as ‘gamma_save’ for ex), with 0 or 1 depending whether the save for the corresponding value of gamma has been done (1) or not (0). Has to be saved, but not to be changed by the user.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
the vertical displacements done to to control (CNL or CNS cases) are in fact damped, through this wallDamping
the height of the upper plate at the very first time step : the engine finds its value []. Has to be saved, but not to be changed by the user.
Class for controlling independently all 6 components of “engineering” stress and strain of periodic Cell. goal are the goal values, while stressMask determines which components prescribe stress and which prescribe strain.
If the strain is prescribed, appropriate strain rate is directly applied. If the stress is prescribed, the strain predictor is used: from stress values in two previous steps the value of strain rate is prescribed so as the value of stress in the next step is as close as possible to the ideal one. Current algorithm is extremly simple and probably will be changed in future, but is roboust enough and mostly works fine.
Stress error (difference between actual and ideal stress) is evaluated in current and previous steps (). Linear extrapolation is used to estimate error in the next step
According to this error, the strain rate is modified by mod parameter
According to this fact, the prescribed stress will (almost) never have exact prescribed value, but the difference would be very small (and decreasing for increasing nSteps. This approach works good if one of the dominant strain rates is prescribed. If all stresses are prescribed or if all goal strains is prescribed as zero, a good estimation is needed for the first step, therefore the compliance matrix is estimated (from user defined estimations of macroscopic material parameters youngEstimation and poissonEstimation) and respective strain rates is computed form prescribed stress rates and compliance matrix (the estimation of compliance matrix could be computed autamatically avoiding user inputs of this kind).
The simulation on rotated periodic cell is also supported. Firstly, the polar decomposition is performed on cell’s transformation matrix trsf , where
is orthogonal (unitary) matrix representing rotation and
is a positive semi-definite Hermitian matrix representing strain. A logarithm of
should be used to obtain realistic values at higher strain values (not implemented yet). A prescribed strain increment in global coordinates
is properly rotated to cell’s local coordinates and added to
The new value of trsf is computed at . From current and next trsf the cell’s velocity gradient velGrad is computed (according to its definition) as
Current implementation allow user to define independent loading “path” for each prescribed component. i.e. define the prescribed value as a function of time (or progress or steps). See Paths.
Examples examples/test/peri3dController_example1.py and examples/test/peri3dController_triaxialCompression.py explain usage and inputs of Peri3dController, examples/test/peri3dController_shear.py is an example of using shear components and also simulation on rotated cell.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Python command (as string) to run when nSteps is achieved. If empty, the engine will be set dead.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Goal state; only the upper triangular matrix is considered; each component is either prescribed stress or strain, depending on stressMask.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Peri3dController internal variable
Peri3dController internal variable
Maximal asolute value of strain allowed in the simulation. If reached, the simulation is considered as finished
Maximal absolute value of strain rate (both normal and shear components of strain)
Predictor modificator, by trail-and-error analysis the value 0.1 was found as the best.
Number of steps of the simulation.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Peri3dController internal variable
Peri3dController internal variable
Peri3dController internal variable
Estimation of macroscopic Poisson’s ratio, used used for the first simulation step
Actual progress of the simulation with Controller.
Peri3dController internal variable
Current strain (deformation) vector (,
,
,
,
,
) (auto-updated).
Peri3dController internal variable
Current strain rate vector.
Current stress vector (,
,
,
,
,
)|yupdate|.
Peri3dController internal variable
Ideal stress vector at current time step.
mask determining whether components of goal are strain (0) or stress (1). The order is 00,11,22,12,02,01 from the least significant bit. (e.g. 0b000011 is stress 00 and stress 11).
Current stress rate vector (that is prescribed, the actual one slightly differ).
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
“Time function” (piecewise linear) for xx direction. Sequence of couples of numbers. First number is time, second number desired value of respective quantity (stress or strain). The last couple is considered as final state (equal to (nSteps, goal)), other values are relative to this state.
Example: nSteps=1000, goal[0]=300, xxPath=((2,3),(4,1),(5,2))
at step 400 (=5*1000/2) the value is 450 (=3*300/2),
at step 800 (=4*1000/5) the value is 150 (=1*300/2),
at step 1000 (=5*1000/5=nSteps) the value is 300 (=2*300/2=goal[0]).
See example scripts/test/peri3dController_example1 for illusration.
Estimation of macroscopic Young’s modulus, used for the first simulation step
Compress/decompress cloud of spheres by controlling periodic cell size until it reaches prescribed average stress, then moving to next stress value in given stress series.
Characteristic length, should be something like mean particle diameter (default -1=invalid value))
Current value of unbalanced force
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Python command to be run when reaching the last specified stress
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
how often to recompute average stress, stiffness and unbalanced force
Exactly keep proportions of the cell (stress is controlled based on average, not its components
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Maximum body span in terms of bbox, to prevent periodic cell getting too small. (auto-computed)
if actual unbalanced force is smaller than this number, the packing is considered stable,
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Current stress value
Where are we at in the stress series
Stresses that should be reached, one after another
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine for independently controlling stress or strain in periodic simulations.
strainStress contains absolute values for the controlled quantity, and stressMask determines meaning of those values (0 for strain, 1 for stress): e.g. ( 1<<0 | 1<<2 ) = 1 | 4 = 5 means that strainStress[0] and strainStress[2] are stress values, and strainStress[1] is strain.
See scripts/test/periodic-triax.py for a simple example.
Absolute stress tolerance
current unbalanced force (updated every globUpdate) (auto-updated)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
python command to be run when the desired state is reached
Imposed stress can be controlled using the packing stiffness or by applying the laws of dynamic (dynCell=true). Don’t forget to assign a mass to the cell.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Work input from boundary controller.
How often to recompute average stress, stiffness and unbalaced force.
Desired stress or strain values (depending on stressMask), strains defined as strain(i)=log(Fii).
Warning
Strains are relative to the O.cell.refSize (reference cell size), not the current one (e.g. at the moment when the new strain value is set).
Damping of cell resizing (0=perfect control, 1=no control at all); see also wallDamping in TriaxialStressController.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
mass of the cell (user set); if not set and dynCell is used, it will be computed as sum of masses of all particles.
maximum body dimension (auto-computed)
Maximum strain rate of the periodic cell.
maximum unbalanced force.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
previous cell grow
Relative stress tolerance
average stiffness (only every globUpdate steps recomputed from interactions) (auto-updated)
cell strain (auto-updated)
cell strain rate (auto-updated)
diagonal terms of the stress tensor
mask determining strain/stress (0/1) meaning for goal components
average stresses, updated at every step (only every globUpdate steps recomputed from interactions if !dynCell)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
The engine perform a triaxial compression with a control in direction ‘i’ in stress (if stressControl_i) else in strain.
For a stress control the imposed stress is specified by ‘sigma_i’ with a ‘max_veli’ depending on ‘strainRatei’. To obtain the same strain rate in stress control than in strain control you need to set ‘wallDamping = 0.8’. For a strain control the imposed strain is specified by ‘strainRatei’. With this engine you can also perform internal compaction by growing the size of particles by using TriaxialStressController::controlInternalStress. For that, just switch on ‘internalCompaction=1’ and fix sigma_iso=value of mean pressure that you want at the end of the internal compaction.
A string appended at the end of all files, use it to name simulations.
mean resultant forces divided by mean contact force
Total packing volume.
current strain rate in direction 1 - converging to ThreeDTriaxialEngine::strainRate1 (./s)
current strain rate in direction 2 - converging to ThreeDTriaxialEngine::strainRate2 (./s)
current strain rate in direction 3 - converging to ThreeDTriaxialEngine::strainRate3 (./s)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
size of the box (2-axis) (auto-updated)
Reference size for strain definition. See TriaxialStressController::depth
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Energy provided by boundaries.|yupdate|
max multiplier of diameters during internal compaction (secondary precise adjustment - TriaxialStressController::maxMultiplier is used in the initial stage)
Value of friction used in the simulation if (updateFrictionAngle)
prescribed stress/strain rate on axis 1, as defined by TriaxialStressController::stressMask (see also TriaxialStressController::isAxisymetric)
prescribed stress/strain rate on axis 2, as defined by TriaxialStressController::stressMask (see also TriaxialStressController::isAxisymetric)
prescribed stress/strain rate on axis 3, as defined by TriaxialStressController::stressMask (see also TriaxialStressController::isAxisymetric)
size of the box (1-axis) (auto-updated)
Reference size for strain definition. See TriaxialStressController::height
Switch between ‘external’ (walls) and ‘internal’ (growth of particles) compaction.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
max multiplier of diameters during internal compaction (initial fast increase - TriaxialStressController::finalMaxMultiplier is used in a second stage)
Maximum allowed walls velocity [m/s]. This value superseeds the one assigned by the stress controller if the later is higher. max_vel can be set to infinity in many cases, but sometimes helps stabilizing packings. Based on this value, different maxima are computed for each axis based on the dimensions of the sample, so that if each boundary moves at its maximum velocity, the strain rate will be isotropic (see e.g. TriaxialStressController::max_vel1).
see TriaxialStressController::max_vel (auto-computed)
see TriaxialStressController::max_vel (auto-computed)
see TriaxialStressController::max_vel (auto-computed)
Mean stress in the packing. (auto-updated)
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Porosity of the packing.
(auto-updated)
(auto-updated)
Assign a new friction angle (degrees) to dynamic bodies and relative interactions
Total volume pf spheres.
target strain rate (./s)
Current strain in a vector (exx,eyy,ezz). The values reflect true (logarithmic) strain.
factor used for smoothing changes in effective strain rate. If target rate is TR, then (1-damping)*(TR-currentRate) will be added at each iteration. With damping=0, rate=target all the time. With damping=1, it doesn’t change.
Current strain rate in a vector d/dt(exx,eyy,ezz).
target strain rate in direction 1 (./s)
target strain rate in direction 2 (./s)
target strain rate in direction 3 (./s)
Return the mean stress vector acting on boundary ‘id’, with ‘id’ between 0 and 5.
Switch to choose a stress or a strain control in directions 1
Switch to choose a stress or a strain control in directions 2
Switch to choose a stress or a strain control in directions 3
wall damping coefficient for the stress control - wallDamping=0 implies a (theoretical) perfect control, wallDamping=1 means no movement
Bitmask determining if the components of TriaxialStressController::goal are stress (1) or strain (0). 0 for none, 7 for all, 1 for sigma1, etc.
thickness of boxes (needed by some functions)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Switch to activate the update of the intergranular frictionto the value ThreeDTriaxialEngine::frictionAngleDegree.
Volumetric strain (see TriaxialStressController::strain).|yupdate|
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 2- (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 1- (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 2+ (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 0- (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 0+ (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 1+ (default value is ok if aabbWalls are appended BEFORE spheres.)
size of the box (0-axis) (auto-updated)
Reference size for strain definition. See TriaxialStressController::width
The engine is a state machine with the following states; transitions my be automatic, see below.
STATE_ISO_COMPACTION: isotropic compaction (compression) until the prescribed mean pressue sigmaIsoCompaction is reached and the packing is stable. The compaction happens either by straining the walls (!internalCompaction) or by growing size of grains (internalCompaction).
STATE_ISO_UNLOADING: isotropic unloading from the previously reached state, until the mean pressure sigmaLateralConfinement is reached (and stabilizes).
Note
this state will be skipped if sigmaLateralConfinement == sigmaIsoCompaction.
STATE_TRIAX_LOADING: confined uniaxial compression: constant sigmaLateralConfinement is kept at lateral walls (left, right, front, back), while top and bottom walls load the packing in their axis (by straining), until the value of epsilonMax (deformation along the loading axis) is reached. At this point, the simulation is stopped.
STATE_FIXED_POROSITY_COMPACTION: isotropic compaction (compression) until a chosen porosity value (parameter:fixedPorosity). The six walls move with a chosen translation speed (parameter StrainRate).
STATE_TRIAX_LIMBO: currently unused, since simulation is hard-stopped in the previous state.
Transition from COMPACTION to UNLOADING is done automatically if autoUnload==true;
Transition from (UNLOADING to LOADING) or from (COMPACTION to LOADING: if UNLOADING is skipped) is done automatically if autoCompressionActivation=true; Both autoUnload and autoCompressionActivation are true by default.
Note
Most of the algorithms used have been developed initialy for simulations reported in [Chareyre2002a] and [Chareyre2005]. They have been ported to Yade in a second step and used in e.g. [Kozicki2008],[Scholtes2009b]_,[Jerier2010b].
A string appended at the end of all files, use it to name simulations.
tolerance in terms of TriaxialCompressionEngine::UnbalancedForce to consider the packing is stable
mean resultant forces divided by mean contact force
Auto-switch from isotropic compaction (or unloading state if sigmaLateralConfinement<sigmaIsoCompaction) to deviatoric loading
Stop the simulation when the sample reach STATE_LIMBO, or keep running
Auto-switch from isotropic compaction to unloading
Total packing volume.
There are 5 possible states in which TriaxialCompressionEngine can be. See above wrapper.TriaxialCompressionEngine
current strain rate - converging to TriaxialCompressionEngine::strainRate (./s)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
size of the box (2-axis) (auto-updated)
Reference size for strain definition. See TriaxialStressController::depth
Return dictionary of attributes.
Value of axial deformation for which the loading must stop
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Energy provided by boundaries.|yupdate|
max multiplier of diameters during internal compaction (secondary precise adjustment - TriaxialStressController::maxMultiplier is used in the initial stage)
A special type of compaction with imposed final porosity TriaxialCompressionEngine::fixedPorosity (WARNING : can give unrealistic results!)
Value of porosity chosen by the user
Value of friction assigned just before the deviatoric loading
prescribed stress/strain rate on axis 1, as defined by TriaxialStressController::stressMask (see also TriaxialStressController::isAxisymetric)
prescribed stress/strain rate on axis 2, as defined by TriaxialStressController::stressMask (see also TriaxialStressController::isAxisymetric)
prescribed stress/strain rate on axis 3, as defined by TriaxialStressController::stressMask (see also TriaxialStressController::isAxisymetric)
size of the box (1-axis) (auto-updated)
Reference size for strain definition. See TriaxialStressController::height
Switch between ‘external’ (walls) and ‘internal’ (growth of particles) compaction.
if true, sigma_iso is assigned to sigma1, 2 and 3 (applies at each iteration and overrides user-set values of s1,2,3)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
max multiplier of diameters during internal compaction (initial fast increase - TriaxialStressController::finalMaxMultiplier is used in a second stage)
Max value of stress during the simulation (for post-processing)
Maximum allowed walls velocity [m/s]. This value superseeds the one assigned by the stress controller if the later is higher. max_vel can be set to infinity in many cases, but sometimes helps stabilizing packings. Based on this value, different maxima are computed for each axis based on the dimensions of the sample, so that if each boundary moves at its maximum velocity, the strain rate will be isotropic (see e.g. TriaxialStressController::max_vel1).
see TriaxialStressController::max_vel (auto-computed)
see TriaxialStressController::max_vel (auto-computed)
see TriaxialStressController::max_vel (auto-computed)
Mean stress in the packing. (auto-updated)
If true, no files will be generated (*.xml, *.spheres,...)
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Porosity of the packing.
(auto-updated)
Previous value of inherited sigma_iso (used to detect manual changes of the confining pressure)
Previous state (used to detect manual changes of the state in .xml)
(auto-updated)
Assign a new friction angle (degrees) to dynamic bodies and relative interactions
Prescribed isotropic pressure during the compaction phase
Prescribed confining pressure in the deviatoric loading; might be different from TriaxialCompressionEngine::sigmaIsoCompaction
prescribed confining stress (see :yref:TriaxialCompressionEngine::isAxisymetric`)
Total volume pf spheres.
target strain rate (./s)
Current strain in a vector (exx,eyy,ezz). The values reflect true (logarithmic) strain.
coefficient used for smoother transitions in the strain rate. The rate reaches the target value like reaches 0, where
is the damping coefficient and
is the number of steps
target strain rate (./s)
Return the mean stress vector acting on boundary ‘id’, with ‘id’ between 0 and 5.
wall damping coefficient for the stress control - wallDamping=0 implies a (theoretical) perfect control, wallDamping=1 means no movement
Bitmask determining if the components of TriaxialStressController::goal are stress (1) or strain (0). 0 for none, 7 for all, 1 for sigma1, etc.
interval of checks for transition between phases, higher than 1 saves computation time.
thickness of boxes (needed by some functions)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
compression axis
Current value of axial deformation during confined loading (is reference to strain[1])
Update object attributes from given dictionary
Volumetric strain (see TriaxialStressController::strain).|yupdate|
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 2- (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 1- (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 2+ (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 0- (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 0+ (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 1+ (default value is ok if aabbWalls are appended BEFORE spheres.)
counter used for sending a deprecation warning once
size of the box (0-axis) (auto-updated)
Reference size for strain definition. See TriaxialStressController::width
An engine maintaining constant stresses or constant strain rates on some boundaries of a parallepipedic packing. The stress/strain control is defined for each axis using TriaxialStressController::stressMask (a bitMask) and target values are defined by goal1,goal2, and goal3. sigmaIso has to be defined during growing phases.
Note
The algorithms used have been developed initialy for simulations reported in [Chareyre2002a] and [Chareyre2005]. They have been ported to Yade in a second step and used in e.g. [Kozicki2008],[Scholtes2009b]_,[Jerier2010b].
Total packing volume.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
size of the box (2-axis) (auto-updated)
Reference size for strain definition. See TriaxialStressController::depth
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Energy provided by boundaries.|yupdate|
max multiplier of diameters during internal compaction (secondary precise adjustment - TriaxialStressController::maxMultiplier is used in the initial stage)
prescribed stress/strain rate on axis 1, as defined by TriaxialStressController::stressMask (see also TriaxialStressController::isAxisymetric)
prescribed stress/strain rate on axis 2, as defined by TriaxialStressController::stressMask (see also TriaxialStressController::isAxisymetric)
prescribed stress/strain rate on axis 3, as defined by TriaxialStressController::stressMask (see also TriaxialStressController::isAxisymetric)
size of the box (1-axis) (auto-updated)
Reference size for strain definition. See TriaxialStressController::height
Switch between ‘external’ (walls) and ‘internal’ (growth of particles) compaction.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
max multiplier of diameters during internal compaction (initial fast increase - TriaxialStressController::finalMaxMultiplier is used in a second stage)
Maximum allowed walls velocity [m/s]. This value superseeds the one assigned by the stress controller if the later is higher. max_vel can be set to infinity in many cases, but sometimes helps stabilizing packings. Based on this value, different maxima are computed for each axis based on the dimensions of the sample, so that if each boundary moves at its maximum velocity, the strain rate will be isotropic (see e.g. TriaxialStressController::max_vel1).
see TriaxialStressController::max_vel (auto-computed)
see TriaxialStressController::max_vel (auto-computed)
see TriaxialStressController::max_vel (auto-computed)
Mean stress in the packing. (auto-updated)
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Porosity of the packing.
(auto-updated)
(auto-updated)
Total volume pf spheres.
target strain rate (./s)
Current strain in a vector (exx,eyy,ezz). The values reflect true (logarithmic) strain.
coefficient used for smoother transitions in the strain rate. The rate reaches the target value like reaches 0, where
is the damping coefficient and
is the number of steps
Current strain rate in a vector d/dt(exx,eyy,ezz).
Return the mean stress vector acting on boundary ‘id’, with ‘id’ between 0 and 5.
wall damping coefficient for the stress control - wallDamping=0 implies a (theoretical) perfect control, wallDamping=1 means no movement
Bitmask determining if the components of TriaxialStressController::goal are stress (1) or strain (0). 0 for none, 7 for all, 1 for sigma1, etc.
thickness of boxes (needed by some functions)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Volumetric strain (see TriaxialStressController::strain).|yupdate|
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 2- (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 1- (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 2+ (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 0- (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 0+ (default value is ok if aabbWalls are appended BEFORE spheres.)
if true, this wall moves according to the target value (stress or strain rate).
id of boundary ; coordinate 1+ (default value is ok if aabbWalls are appended BEFORE spheres.)
size of the box (0-axis) (auto-updated)
Reference size for strain definition. See TriaxialStressController::width
Axial displacing two groups of bodies in the opposite direction with given strain rate.
alternatively, absolute speed of boundary motion can be specified; this is effective only at the beginning and if strainRate is not set; changing absSpeed directly during simulation wil have no effect. [ms⁻¹]
Whether this engine is activated
If 0, straining is symmetric for negIds and posIds; for 1 (or -1), only posIds are strained and negIds don’t move (or vice versa)
Current average stress (auto-updated) [Pa]
The axis which is strained (0,1,2 for x,y,z)
Whether displacement of boundary bodies perpendicular to the strained axis are blocked of are free
Whether rotations of boundary bodies are blocked.
crossSection perpendicular to he strained axis; must be given explicitly [m²]
Current strain rate (update automatically). (auto-updated)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Number of iterations that will pass without straining activity after stopStrain has been reached
Time for strain reaching the requested value (linear interpolation). If negative, the time is dt*(-initAccelTime), where dt is the timestep at the first iteration. [s]
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Invert the sense of straining (sharply, without transition) one this value of strain is reached. Not effective if 0.
Bodies on which strain will be applied (on the negative end along the axis)
Flag whether the sense of straining has already been reversed (only used internally).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Distance of reference bodies in the direction of axis before straining started (computed automatically) [m]
Bodies on which strain will be applied (on the positive end along the axis)
should we set speeds at the beginning directly, instead of increasing strain rate progressively?
Strain at which we will pause simulation; inactive (nan) by default; must be reached from below (in absolute value)
Current strain value, elongation/originalLength (auto-updated) [-]
Rate of strain, starting at 0, linearly raising to strainRate. [-]
How often to recompute stress on supports.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Abstract class for finding spatial collisions between bodies.
Special constructor
Derived colliders (unless they override pyHandleCustomCtorArgs) can be given list of BoundFunctors which is used to initialize the internal boundDispatcher instance.
This mask is used to avoid the interactions inside a group of particles. To do so, the particles must have the same mask and this mask have to be compatible with this one.
BoundDispatcher object that is used for creating bounds on collider’s request as necessary.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Non-optimized grid collider, storing grid as dense flat array. Each body is assigned to (possibly multiple) cells, which are arranged in regular grid between aabbMin and aabbMax, with cell size step (same in all directions). Bodies outsize (aabbMin, aabbMax) are handled gracefully, assigned to closest cells (this will create spurious potential interactions). verletDist determines how much is each body enlarged to avoid collision detection at every step.
Note
This collider keeps all cells in linear memory array, therefore will be memory-inefficient for sparse simulations.
Warning
objects Body::bound are not used, BoundFunctors are not used either: assigning cells to bodies is hard-coded internally. Currently handles Shapes are: Sphere.
Note
Periodic boundary is not handled (yet).
Upper corner of grid (approximate, might be rouded up to minStep.
Lower corner of grid.
This mask is used to avoid the interactions inside a group of particles. To do so, the particles must have the same mask and this mask have to be compatible with this one.
BoundDispatcher object that is used for creating bounds on collider’s request as necessary.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Step in the grid (cell size)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Length by which enlarge space occupied by each particle; avoids running collision detection at every step.
Collider with O(n log(n)) complexity, using Aabb for bounds.
At the initial step, Bodies’ bounds (along sortAxis) are first std::sort’ed along one axis (sortAxis), then collided. The initial sort has complexity, see Colliders’ performance for some information (There are scripts in examples/collider-perf for measurements).
Insertion sort is used for sorting the bound list that is already pre-sorted from last iteration, where each inversion calls checkOverlap which then handles either overlap (by creating interaction if necessary) or its absence (by deleting interaction if it is only potential).
Bodies without bounding volume (such as clumps) are handled gracefully and never collide. Deleted bodies are handled gracefully as well.
This collider handles periodic boundary conditions. There are some limitations, notably:
- No body can have Aabb larger than cell’s half size in that respective dimension. You get exception if it does and gets in interaction. One way to explicitly by-pass this restriction is offered by allowBiggerThanPeriod, which can be turned on to insert a floor in the form of a very large box for instance (see examples/periodicSandPile.py).
- No body can travel more than cell’s distance in one step; this would mean that the simulation is numerically exploding, and it is only detected in some cases.
Stride can be used to avoid running collider at every step by enlarging the particle’s bounds, tracking their displacements and only re-run if they might have gone out of that bounds (see Verlet list for brief description and background) . This requires cooperation from NewtonIntegrator as well as BoundDispatcher, which will be found among engines automatically (exception is thrown if they are not found).
If you wish to use strides, set verletDist (length by which bounds will be enlarged in all directions) to some value, e.g. 0.05 × typical particle radius. This parameter expresses the tradeoff between many potential interactions (running collider rarely, but with longer exact interaction resolution phase) and few potential interactions (running collider more frequently, but with less exact resolutions of interactions); it depends mainly on packing density and particle radius distribution.
If targetInterv is >1, not all particles will have their bound enlarged by verletDist; instead, they will have bounds increased by a length in order to trigger a new colliding after targetInterv iteration, assuming they move at almost constant velocity. Ideally in this method, all particles would reach their bounds at the sime iteration. This is of course not the case as soon as velocities fluctuate in time. Bound::sweepLength is tuned on the basis of the displacement recorded between the last two runs of the collider. In this situation, verletDist defines the maximum sweep length.
If true, tests on bodies sizes will be disabled, and the simulation will run normaly even if bodies larger than period are found. It can be useful when the periodic problem include e.g. a floor modelized with wall/box/facet. Be sure you know what you are doing if you touch this flag. The result is undefined if one large body moves out of the (0,0,0) period.
This mask is used to avoid the interactions inside a group of particles. To do so, the particles must have the same mask and this mask have to be compatible with this one.
BoundDispatcher object that is used for creating bounds on collider’s request as necessary.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Do forced resorting of interactions.
Return representation of the internal sort data. The format is ([...],[...],[...]) for 3 axes, where each ... is a list of entries (bounds). The entry is a tuple with the fllowing items:
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Normalized maximum displacement of the fastest body since last run; if >= 1, we could get out of bboxes and will trigger full run. (auto-updated)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Minimal distance by which enlarge all bounding boxes; superseeds computed value of verletDist when lower that (minSweepDistFactor x verletDist).
Cummulative number of bound array re-initialization.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Whether the collider is in periodic mode (read-only; for debugging) (auto-updated)
Axis for the initial contact detection.
Separate sorting and colliding phase; it is MUCH slower, but all interactions are processed at every step; this effectively makes the collider non-persistent, not remembering last state. (The default behavior relies on the fact that inversions during insertion sort are overlaps of bounding boxes that just started/ceased to exist, and only processes those; this makes the collider much more efficient.)
Whether striding is active (read-only; for debugging). (auto-updated)
(experimental) Target number of iterations between bound update, used to define a smaller sweep distance for slower grains if >0, else always use 1*verletDist. Useful in simulations with strong velocity contrasts between slow bodies and fast bodies.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
(experimental) Displacement factor used to trigger bound update: the bound is updated only if updatingDispFactor*disp>sweepDist when >0, else all bounds are updated.
for compatibility of scripts defining the old collider’s attributes - see deprecated attributes
Length by which to enlarge particle bounds, to avoid running collider at every step. Stride disabled if zero. Negative value will trigger automatic computation, so that the real value will be verletDist × minimum spherical particle radius; if there are no spherical particles, it will be disabled. The actual length added to one bound can be only a fraction of verletDist when InsertionSortCollider::targetInterv is > 0.
Collision detection engine based on regular triangulation. Handles spheres and flat boundaries (considered as infinite-sized bounding spheres).
This mask is used to avoid the interactions inside a group of particles. To do so, the particles must have the same mask and this mask have to be compatible with this one.
BoundDispatcher object that is used for creating bounds on collider’s request as necessary.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Keep distant interactions? If True, don’t delete interactions once bodies don’t overlap anymore; constitutive laws will be responsible for requesting deletion. If False, delete as soon as there is no object penetration.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Collider using quicksort along axes at each step, using Aabb bounds.
Its performance is lower than that of InsertionSortCollider (see Colliders’ performance), but the algorithm is simple enought to make it good for checking other collider’s correctness.
This mask is used to avoid the interactions inside a group of particles. To do so, the particles must have the same mask and this mask have to be compatible with this one.
BoundDispatcher object that is used for creating bounds on collider’s request as necessary.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Collider with O(n log(n)) complexity, using a CGAL algorithm from Zomorodian and Edelsbrunner [Kettner2011] (http://www.cgal.org/Manual/beta/doc_html/cgal_manual/Box_intersection_d/Chapter_main.html)
This mask is used to avoid the interactions inside a group of particles. To do so, the particles must have the same mask and this mask have to be compatible with this one.
BoundDispatcher object that is used for creating bounds on collider’s request as necessary.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Maximum displacement of the fastest body since last run; if >= verletDist, we could get out of bboxes and will trigger full run. DEPRECATED, was only used without bins. (auto-updated)
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Cummulative number of bound array re-initialization.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Whether the collider is in periodic mode (read-only; for debugging) (auto-updated)
Axis for the initial contact detection.
Separate sorting and colliding phase; it is MUCH slower, but all interactions are processed at every step; this effectively makes the collider non-persistent, not remembering last state. (The default behavior relies on the fact that inversions during insertion sort are overlaps of bounding boxes that just started/ceased to exist, and only processes those; this makes the collider much more efficient.)
Whether striding is active (read-only; for debugging). (auto-updated)
(experimental) Target number of iterations between bound update, used to define a smaller sweep distance for slower grains if >0, else always use 1*verletDist. Useful in simulations with strong velocity contrasts between slow bodies and fast bodies.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
(experimental) Displacement factor used to trigger bound update: the bound is updated only if updatingDispFactor*disp>sweepDist when >0, else all bounds are updated.
Length by which to enlarge particle bounds, to avoid running collider at every step. Stride disabled if zero. Negative value will trigger automatic computation, so that the real value will be verletDist × minimum spherical particle radius; if there are no spherical particles, it will be disabled.
Base for engines applying force files on particles. Not to be used directly.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Apply acceleration (independent of distance) directed towards an axis.
Acceleration magnitude [kgms⁻²]
direction of the gravity axis (will be normalized automatically)
Point through which the axis is passing.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
If mask defined, only bodies with corresponding groupMask will be affected by this engine. If 0, all bodies will be affected.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine applying acceleration to all bodies, towards a central body.
Acceleration magnitude [kgms⁻²]
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
If mask defined, only bodies with corresponding groupMask will be affected by this engine. If 0, all bodies will be affected.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
If true, acceleration will be applied on the central body as well.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine applying constant acceleration to all bodies. DEPRECATED, use Newton::gravity unless you need energy tracking or selective gravity application using groupMask).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Acceleration [kgms⁻²]
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
If mask defined, only bodies with corresponding groupMask will be affected by this engine. If 0, all bodies will be affected.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
For deprecation warning once.
Read accelerometer in Thinkpad laptops (HDAPS and accordingly set gravity within the simulation. This code draws from hdaps-gl . See scripts/test/hdaps.py for an example.
reading from the sysfs file
Zero position; if NaN, will be read from the hdapsDir / calibrate.
Whether calibrate was already updated. Do not set to True by hand unless you also give a meaningful value for calibrate.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Acceleration [kgms⁻²]
Hdaps directory; contains position (with accelerometer readings) and calibration (zero acceleration).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
If mask defined, only bodies with corresponding groupMask will be affected by this engine. If 0, all bodies will be affected.
How often to update the reading.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Minimum difference of reading from the file before updating gravity, to avoid jitter.
For deprecation warning once.
Gravity if the accelerometer is in flat (zero) position.
Engine affecting only particular bodies in the simulation, defined by ids.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine applying the linear motion of bicycle pedal e.g. moving points around the axis without rotation
Angular velocity. [rad/s]
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Initial phase [radians]
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Rotation radius. [m]
Axis of rotation (direction); will be normalized automatically.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine for applying combined displacements on pre-defined bodies. Constructed using + operator on regular KinematicEngines. The ids operated on are those of the first engine in the combination (assigned automatically).
Kinematic engines that will be combined by this one, run in the order given.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Apply drag force on some particles at each step, decelerating them proportionally to their linear velocities. The applied force reads
where is the medium density (density),
is particle’s velocity,
is particle projected area (disc),
is the drag coefficient (0.47 for Sphere),
Note
Drag force is only applied to spherical particles, listed in ids.
Drag coefficient <http://en.wikipedia.org/wiki/Drag_coefficient>`_.
Density of the medium (fluid or air), by default - the density of the air.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
An engine to solve flow problem in saturated granular media. Model description can be found in [Chareyre2012a] and [Catalano2013a]. See the example script FluidCouplingPFV/oedometer.py. More documentation to come.
Activate debug messages
The value used for initializing pore pressure. It is useless for incompressible fluid, but important for compressible model.
Gauss-Seidel relaxation
Gauss-Seidel Tolerance
Allow sinusoidal pressure condition to simulate ocean waves
Compute a sphere-centered average fluid velocity
Measure averaged pore pressure in the entire volume
Measure slice-averaged pore pressure at height posY
defines the type of boundary condition for each side. True if pressure is imposed, False for no-flux. Indexes can be retrieved with FlowEngine::xmin and friends.
Imposed value of a boundary condition. Only applies if the boundary condition is imposed pressure, else the imposed flux is always zero presently (may be generalized to non-zero imposed fluxes in the future).
Return the normal lubrication stress on sphere Id_sph.
Return the shear lubrication stress on sphere Id_sph.
If true (default value) bounding sphere is added as function of max/min sphere coord, if false as function of yade wall position
velocity on top boundary, only change it using FlowEngine::setBoundaryVel
get statistics of cholmod solver activity
If true, clamp local permeabilities in [minKdivKmean,maxKdivKmean]*globalK. This clamping can avoid singular values in the permeability matrix and may reduce numerical errors in the solve phase. It will also hide junk values if they exist, or bias all values in very heterogeneous problems. So, use this with care.
Clear the list of points with flux imposed.
Clear the list of points with pressure imposed.
create file of velocities
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Cumulated deformation threshold for which retriangulation of pore space is performed. If negative, the triangulation update will occure with a fixed frequency on the basis of FlowEngine::meshUpdateInterval
Return dictionary of attributes.
display the lubrication force applied on particles
timestep [s]
get scene and run action (may be used to manipulate an engine outside the timestepping loop).
roughness defined as a fraction of particles size, giving the minimum distance between particles in the lubrication model.
Maximal absolute volumetric strain computed at each iteration. (auto-updated)
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Export system matrix to a file with all entries (even zeros will displayed).
Export system matrix to a file with only non-zero entries.
Controls the initialization/update phases
Bulk modulus of fluid (inverse of compressibility) K=-dP*V/dV [Pa]. Flow is compressible if fluidBulkModulus > 0, else incompressible.
Return the fluid force on sphere Id_sph.
If true, METIS is used for matrix preconditioning, else Cholmod is free to choose the best method (which may be METIS to, depending on the matrix). See nmethods in Cholmod documentation
Get total flux through boundary defined by its body id.
Note
The flux may be not zero even for no-flow condition. This artifact comes from cells which are incident to two or more boundaries (along the edges of the sample, typically). Such flux evaluation on impermeable boundary is just irrelevant, it does not imply that the boundary condition is not applied properly.
get id of the cell containing (X,Y,Z).
Get influx in cell associated to an imposed P (indexed using ‘cond’).
Get the list of constriction radii (inscribed circle) for all finite facets (if all==True) or all facets not incident to a virtual bounding sphere (if all==False). When all facets are returned, negative radii denote facet incident to one or more fictious spheres.
Get the list of constrictions (inscribed circle) for all finite facets (if all==True), or all facets not incident to a fictious bounding sphere (if all==False). When all facets are returned, negative radii denote facet incident to one or more fictious spheres. The constrictions are returned in the format {{cell1,cell2}{rad,nx,ny,nz}}
Measure pore pressure in position pos[0],pos[1],pos[2]
get the vertices of a cell
Id of a sphere to exclude from the triangulation.)
Impose incoming flux in boundary cell of location ‘pos’.
Impose pressure in cell of location ‘pos’. The index of the condition is returned (for multiple imposed pressures at different points).
Activates Flow Engine
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
define the max K value (see FlowEngine::clampKValues)
report the local permeabilities’ correction
Maximum number of timesteps between re-triangulation events. See also FlowEngine::defTolerance.
check wether metis lib is effectively used
define the min K value (see FlowEngine::clampKValues)
Build triangulation and factorize in the background (multi-thread mode)
get the total number of finite cells in the triangulation.
Return the normal lubrication force on sphere Id_sph.
Compute normal lubrication force as developped by Brule
Return the normal vector between particles.
Return the normal velocity of the interaction.
number of openblas threads in the factorization phase
number of openblas threads in the solve phase.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Return the id of the interaction only between spheres.
permability multiplier
Enable/disable stocking of average permeability scalar in cell infos.
Porosity computed at each retriangulation (auto-updated)
Compute the pressure field and associated fluid forces. WARNING: turning off means fluid flow is not computed at all.
Measure pore pressure in 6 equally-spaced points along the height of the sample
Save pressure field in vtk format.
Set pressure value at the point indexed ‘cond’.
Return the shear lubrication force on sphere Id_sph.
Return the shear lubrication torque on sphere Id_sph.
Compute shear lubrication force as developped by Brule (FIXME: ref.)
Return the shear velocity of the interaction.
Pressure value (average) when sinusoidal pressure is applied
Pressure value (amplitude) when sinusoidal pressure is applied (p )
Controls friction condition on lateral walls
equivalent contact stiffness used in the lubrication model
Return the distance between particles.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
tells the engine to update it’s boundary conditions before running (especially useful when changing boundary pressure - should not be needed for point-wise imposed pressure)
If true the medium is retriangulated. Can be switched on to force retriangulation after some events (else it will be true periodicaly based on FlowEngine::defTolerance and FlowEngine::meshUpdateInterval. Of course, it costs CPU time.
Solver to use 0=G-Seidel, 1=Taucs, 2-Pardiso, 3-CHOLMOD
viscosity of the fluid
Compute normal viscous stress applied on each body
Compute viscous shear terms as developped by Donia Marzougui (FIXME: ref.)
Compute shear viscous stress applied on each body
body ids of the boundaries (default values are ok only if aabbWalls are appended before spheres, i.e. numbered 0,...,5)
Walls thickness
See FlowEngine::xmin.
Index of the boundary . This index is not equal the the id of the corresponding body in general, it may be used to access the corresponding attributes (e.g. flow.bndCondValue[flow.xmin], flow.wallId[flow.xmin],...).
See FlowEngine::xmin.
See FlowEngine::xmin.
See FlowEngine::xmin.
See FlowEngine::xmin.
Apply contact force on some particles at each step.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Force to apply.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
This engine implements the harmonic oscillation of bodies. http://en.wikipedia.org/wiki/Simple_harmonic_motion#Dynamics_of_simple_harmonic_motion
Amplitude [m]
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Frequency [hertz]
Initial phase [radians]. By default, the body oscillates around initial position.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
This engine implements the harmonic-rotation oscillation of bodies. http://en.wikipedia.org/wiki/Simple_harmonic_motion#Dynamics_of_simple_harmonic_motion ; please, set dynamic=False for bodies, droven by this engine, otherwise amplitude will be 2x more, than awaited.
Amplitude [rad]
Angular velocity. [rad/s]
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Frequency [hertz]
Initial phase [radians]. By default, the body oscillates around initial position.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
If True, bodies will not rotate around their centroids, but rather around zeroPoint.
Axis of rotation (direction); will be normalized automatically.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Point around which bodies will rotate if rotateAroundZero is True
Engine applying both rotation and translation, along the same axis, whence the name HelixEngine
How much have we turned so far. (auto-updated) [rad]
Angular velocity. [rad/s]
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Linear velocity [m/s]
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
If True, bodies will not rotate around their centroids, but rather around zeroPoint.
Axis of rotation (direction); will be normalized automatically.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Point around which bodies will rotate if rotateAroundZero is True
Engine for applying force of varying magnitude but constant direction on subscribed bodies. times and magnitudes must have the same length, direction (normalized automatically) gives the orientation.
As usual with interpolating engines: the first magnitude is used before the first time point, last magnitude is used after the last time point. Wrap specifies whether time wraps around the last time point to the first time point.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Contact force direction (normalized automatically)
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Force to apply.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Force magnitudes readings [N]
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Time readings [s]
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
wrap to the beginning of the sequence if beyond the last time point
Engine applying spiral motion, finding current angular velocity by linearly interpolating in times and velocities and translation by using slope parameter.
The interpolation assumes the margin value before the first time point and last value after the last time point. If wrap is specified, time will wrap around the last times value to the first one (note that no interpolation between last and first values is done).
How much have we turned so far. (auto-updated) [rad]
List of angular velocities; manadatorily of same length as times. [rad/s]
Angular velocity. [rad/s]
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Linear velocity [m/s]
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
If True, bodies will not rotate around their centroids, but rather around zeroPoint.
Axis of rotation (direction); will be normalized automatically.
Axial translation per radian turn (can be negative) [m/rad]
List of time points at which velocities are given; must be increasing [s]
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Wrap t if t>times_n, i.e. t_wrapped=t-N*(times_n-times_0)
Point around which bodies will rotate if rotateAroundZero is True
Abstract engine for applying prescribed displacement.
Note
Derived classes should override the apply with given list of ids (not action with PartialEngine.ids), so that they work when combined together; velocity and angular velocity of all subscribed bodies is reset before the apply method is called, it should therefore only increment those quantities.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Prescribe and apply deformations of an interaction in terms of local mutual displacements and rotations. The loading path is specified either using path (as sequence of 6-vectors containing generalized displacements ,
,
,
,
,
) or disPath (
,
,
) and rotPath (
,
,
). Time function with time values (step numbers) corresponding to points on loading path is given by pathSteps. Loading values are linearly interpolated between given loading path points, and starting zero-value (the initial configuration) is assumed for both path and pathSteps. hooks can specify python code to run when respective point on the path is reached; when the path is finished, doneHook will be run.
LawTester should be placed between InteractionLoop and NewtonIntegrator in the simulation loop, since it controls motion via setting linear/angular velocities on particles; those velocities are integrated by NewtonIntegrator to yield an actual position change, which in turn causes IGeom to be updated (and contact law applied) when InteractionLoop is executed. Constitutive law generating forces on particles will not affect prescribed particle motion, since both particles have all DoFs blocked when first used with LawTester.
LawTester uses, as much as possible, IGeom to provide useful data (such as local coordinate system), but is able to compute those independently if absent in the respective IGeom:
IGeom | #DoFs | LawTester support level |
---|---|---|
L3Geom | 3 | full |
L6Geom | 6 | full |
ScGeom | 3 | emulate local coordinate system |
ScGeom6D | 6 | emulate local coordinate system |
Depending on IGeom, 3 (,
,
) or 6 (
,
,
,
,
,
) degrees of freedom (DoFs) are controlled with LawTester, by prescribing linear and angular velocities of both particles in contact. All DoFs controlled with LawTester are orthogonal (fully decoupled) and are controlled independently.
When 3 DoFs are controlled, rotWeight controls whether local shear is applied by moving particle on arc around the other one, or by rotating without changing position; although such rotation induces mutual rotation on the interaction, it is ignored with IGeom with only 3 DoFs. When 6 DoFs are controlled, only arc-displacement is applied for shear, since otherwise mutual rotation would occur.
idWeight distributes prescribed motion between both particles (resulting local deformation is the same if id1 is moved towards id2 or id2 towards id1). This is true only for ,
,
,
however ; bending rotations
,
are nevertheless always distributed regardless of idWeight to both spheres in inverse proportion to their radii, so that there is no shear induced.
LawTester knows current contact deformation from 2 sources: from its own internal data (which are used for prescribing the displacement at every step), which can be accessed in uTest, and from IGeom itself (depending on which data it provides), which is stored in uGeom. These two values should be identical (disregarding numerical percision), and it is a way to test whether IGeom and related functors compute what they are supposed to compute.
LawTester-operated interactions can be rendered with GlExtra_LawTester renderer.
See scripts/test/law-test.py for an example.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Loading path, where each Vector3 contains desired normal displacement and two components of the shear displacement (in local coordinate system, which is being tracked automatically. If shorter than rotPath, the last value is repeated.
Whether displacement values in disPath are normalized by reference contact length (r1+r2 for 2 spheres).
Python command (as string) to run when end of the path is achieved. If empty, the engine will be set dead.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Python commands to be run when the corresponding point in path is reached, before doing other things in that particular step. See also doneHook.
Float, usually ∈〈0,1〉, determining on how are displacements distributed between particles (0 for id1, 1 for id2); intermediate values will apply respective part to each of them. This parameter is ignored with 6-DoFs IGeom.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Step number for corresponding values in path; if shorter than path, distance between last 2 values is used for the rest.
Reference contact length, for rendering only.
Characteristic length for the purposes of rendering, set equal to the smaller radius.
Rotational components of the loading path, where each item contains torsion and two bending rotations in local coordinates. If shorter than path, the last value is repeated.
Float ∈〈0,1〉 determining whether shear displacement is applied as rotation or displacement on arc (0 is displacement-only, 1 is rotation-only). Not effective when mutual rotation is specified.
Step number in which this engine is active; determines position in path, using pathSteps.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Transformation matrix for the local coordinate system. (auto-updated)
Current generalized displacements (3 displacements, 3 rotations), as stored in the interation itself. They should corredpond to uTest, otherwise a bug is indicated.
Current generalized displacements (3 displacements, 3 rotations), as they should be according to this LawTester. Should correspond to uGeom.
Update object attributes from given dictionary
Generalized displacement values reached in the previous step, for knowing which increment to apply in the current step.
Apply viscous resistance or linear drag on some particles at each step, decelerating them proportionally to their linear velocities. The applied force reads
where is the linear drag,
is particle’s velocity.
where is the medium viscosity,
is the Stokes radius of the particle (but in this case we accept it equal to sphere radius for simplification),
Note
linear drag is only applied to spherical particles, listed in ids.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Viscosity of the medium.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
A variant of FlowEngine implementing periodic boundary conditions. The API is very similar.
Activate debug messages
Return the number of interactions only between spheres.
The value used for initializing pore pressure. It is useless for incompressible fluid, but important for compressible model.
Gauss-Seidel relaxation
Gauss-Seidel Tolerance
Allow sinusoidal pressure condition to simulate ocean waves
Compute a sphere-centered average fluid velocity
Measure averaged pore pressure in the entire volume
Measure slice-averaged pore pressure at height posY
defines the type of boundary condition for each side. True if pressure is imposed, False for no-flux. Indexes can be retrieved with FlowEngine::xmin and friends.
Imposed value of a boundary condition. Only applies if the boundary condition is imposed pressure, else the imposed flux is always zero presently (may be generalized to non-zero imposed fluxes in the future).
Return the normal lubrication stress on sphere Id_sph.
Return the shear lubrication stress on sphere Id_sph.
If true (default value) bounding sphere is added as function of max/min sphere coord, if false as function of yade wall position
velocity on top boundary, only change it using FlowEngine::setBoundaryVel
get statistics of cholmod solver activity
If true, clamp local permeabilities in [minKdivKmean,maxKdivKmean]*globalK. This clamping can avoid singular values in the permeability matrix and may reduce numerical errors in the solve phase. It will also hide junk values if they exist, or bias all values in very heterogeneous problems. So, use this with care.
Clear the list of points with flux imposed.
Clear the list of points with pressure imposed.
Like TesselationWrapper::computeVolumes()
create file of velocities
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Cumulated deformation threshold for which retriangulation of pore space is performed. If negative, the triangulation update will occure with a fixed frequency on the basis of FlowEngine::meshUpdateInterval
Return dictionary of attributes.
display the lubrication force applied on particles
timestep [s]
distance from cell borders that will triger periodic duplication in the triangulation (auto-updated)
Return the number of interactions.
get scene and run action (may be used to manipulate an engine outside the timestepping loop).
roughness defined as a fraction of particles size, giving the minimum distance between particles in the lubrication model.
Maximal absolute volumetric strain computed at each iteration. (auto-updated)
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Export system matrix to a file with all entries (even zeros will displayed).
Export system matrix to a file with only non-zero entries.
Controls the initialization/update phases
Bulk modulus of fluid (inverse of compressibility) K=-dP*V/dV [Pa]. Flow is compressible if fluidBulkModulus > 0, else incompressible.
Return the fluid force on sphere Id_sph.
If true, METIS is used for matrix preconditioning, else Cholmod is free to choose the best method (which may be METIS to, depending on the matrix). See nmethods in Cholmod documentation
Get total flux through boundary defined by its body id.
get id of the cell containing ‘pos’.
Get influx in cell associated to an imposed P (indexed using ‘cond’).
Get the list of constriction radii (inscribed circle) for all finite facets (if all==True) or all facets not incident to a virtual bounding sphere (if all==False). When all facets are returned, negative radii denote facet incident to one or more fictious spheres.
Get the list of constrictions (inscribed circle) for all finite facets (if all==True), or all facets not incident to a fictious bounding sphere (if all==False). When all facets are returned, negative radii denote facet incident to one or more fictious spheres. The constrictions are returned in the format {{cell1,cell2}{rad,nx,ny,nz}}
Measure pore pressure in position pos[0],pos[1],pos[2]
get the vertices of a cell
Macroscopic pressure gradient
Id of a sphere to exclude from the triangulation.)
Impose incoming flux in boundary cell of location ‘pos’.
Impose pressure in cell of location ‘pos’. The index of the condition is returned (for multiple imposed pressures at different points).
Activates Flow Engine
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
define the max K value (see FlowEngine::clampKValues)
report the local permeabilities’ correction
measure the mean velocity in the period
Maximum number of timesteps between re-triangulation events. See also FlowEngine::defTolerance.
check wether metis lib is effectively used
define the min K value (see FlowEngine::clampKValues)
Build triangulation and factorize in the background (multi-thread mode)
get the total number of finite cells in the triangulation.
Return the normal lubrication force on sphere Id_sph.
Compute normal lubrication force as developped by Brule
Return the shear velocity of the interaction.
Return the normal vector between particles.
Return the normal velocity of the interaction.
number of openblas threads in the factorization phase
number of openblas threads in the solve phase.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Return the id of the interaction only between spheres.
permability multiplier
Enable/disable stocking of average permeability scalar in cell infos.
Porosity computed at each retriangulation (auto-updated)
Compute the pressure field and associated fluid forces. WARNING: turning off means fluid flow is not computed at all.
Measure pore pressure in 6 equally-spaced points along the height of the sample
Save pressure field in vtk format.
Set pressure value at the point indexed ‘cond’.
Return the shear lubrication force on sphere Id_sph.
Return the shear lubrication torque on sphere Id_sph.
Compute shear lubrication force as developped by Brule (FIXME: ref.)
Return the shear velocity of the interaction.
Return the shear velocity of the interaction.
Pressure value (average) when sinusoidal pressure is applied
Pressure value (amplitude) when sinusoidal pressure is applied (p )
Controls friction condition on lateral walls
equivalent contact stiffness used in the lubrication model
Return the distance between particles.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
tells the engine to update it’s boundary conditions before running (especially useful when changing boundary pressure - should not be needed for point-wise imposed pressure)
If true the medium is retriangulated. Can be switched on to force retriangulation after some events (else it will be true periodicaly based on FlowEngine::defTolerance and FlowEngine::meshUpdateInterval. Of course, it costs CPU time.
Solver to use 0=G-Seidel, 1=Taucs, 2-Pardiso, 3-CHOLMOD
viscosity of the fluid
Compute normal viscous stress applied on each body
Compute viscous shear terms as developped by Donia Marzougui (FIXME: ref.)
Compute shear viscous stress applied on each body
Returns the volume of Voronoi’s cell of a sphere.
body ids of the boundaries (default values are ok only if aabbWalls are appended before spheres, i.e. numbered 0,...,5)
Walls thickness
See FlowEngine::xmin.
Index of the boundary . This index is not equal the the id of the corresponding body in general, it may be used to access the corresponding attributes (e.g. flow.bndCondValue[flow.xmin], flow.wallId[flow.xmin],...).
See FlowEngine::xmin.
See FlowEngine::xmin.
See FlowEngine::xmin.
See FlowEngine::xmin.
Apply force of given magnitude directed away from spatial axis.
Axis direction (normalized automatically)
Point on axis
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Applied force magnitude
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine applying rotation (by setting angular velocity) to subscribed bodies. If rotateAroundZero is set, then each body is also displaced around zeroPoint.
Angular velocity. [rad/s]
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
If True, bodies will not rotate around their centroids, but rather around zeroPoint.
Axis of rotation (direction); will be normalized automatically.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Point around which bodies will rotate if rotateAroundZero is True
PIDController servo-engine for applying prescribed force on bodies. http://en.wikipedia.org/wiki/PID_controller
Unit vector along which apply the velocity [-]
Current applied velocity [m/s]
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Current error [N]
Previous error [N]
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Integral term [N]
Periodicity criterion of velocity correlation [-]
Previous iteration of velocity correlation [-]
Derivative gain/coefficient for the PID-controller [-]
Integral gain/coefficient for the PID-controller [-]
Proportional gain/coefficient for the PID-controller [-]
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Velocity [m/s]
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Target value for the controller [N]
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Direction [Vector3]
Update object attributes from given dictionary
Velocity [m/s]
Apply generalized displacement (displacement or rotation) stepwise on subscribed bodies. Could be used for purposes of contact law tests (by moving one sphere compared to another), but in this case, see rather LawTester
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Linear displacement step to be applied per iteration, by addition to State.pos.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Rotation step to be applied per iteration (via rotation composition with State.ori).
If false, positions and orientations are directly updated, without changing the speeds of concerned bodies. If true, only velocity and angularVelocity are modified. In this second case integrator is supposed to be used, so that, thanks to this Engine, the bodies will have the prescribed jump over one iteration (dt).
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Apply given torque (momentum) value at every subscribed particle, at every step.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Torque value to be applied.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
This engine is the base class for different engines, which require any kind of motion.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Direction [Vector3]
Update object attributes from given dictionary
Velocity [m/s]
Functor for creating/updating Body::bound.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create/update an Aabb of a Box.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Functor creating Aabb from ChainedCylinder.
Relative enlargement of the bounding box; deactivated if negative.
Note
This attribute is used to create distant interaction, but is only meaningful with an IGeomFunctor which will not simply discard such interactions: Ig2_Cylinder_Cylinder_ScGeom::interactionDetectionFactor should have the same value as aabbEnlargeFactor.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Functor creating Aabb from Cylinder.
Relative enlargement of the bounding box; deactivated if negative.
Note
This attribute is used to create distant interaction, but is only meaningful with an IGeomFunctor which will not simply discard such interactions: Ig2_Cylinder_Cylinder_ScGeom::interactionDetectionFactor should have the same value as aabbEnlargeFactor.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Creates/updates an Aabb of a Facet.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Functor creating Aabb from a GridConnection.
Relative enlargement of the bounding box; deactivated if negative.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create/update Aabb of a Polyhedra
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Functor creating Aabb from Sphere.
Relative enlargement of the bounding box; deactivated if negative.
Note
This attribute is used to create distant interaction, but is only meaningful with an IGeomFunctor which will not simply discard such interactions: Ig2_Sphere_Sphere_ScGeom::interactionDetectionFactor should have the same value as aabbEnlargeFactor.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Creates/updates an Aabb of a Wall
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Dispatcher calling functors based on received argument type(s).
Whether the engine is activated (only should be changed by the collider)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Return functor that would be dispatched for given argument(s); None if no dispatch; ambiguous dispatch throws.
Return dictionary with contents of the dispatch matrix.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Functors associated with this dispatcher.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Minimal distance by which enlarge all bounding boxes; superseeds computed value of sweepDist when lower that (minSweepDistFactor x sweepDist). Updated by the collider. (auto-updated).
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Distance by which enlarge all bounding boxes, to prevent collider from being run at every step (only should be changed by the collider).
see InsertionSortCollider::targetInterv (auto-updated)
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
see InsertionSortCollider::updatingDispFactor (auto-updated)
Functor for creating/updating Interaction::geom objects.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create an interaction geometry ScGeom from Box and Sphere, representing the box with a projected virtual sphere of same radius.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create an interaction geometry ScGeom6D from Box and Sphere, representing the box with a projected virtual sphere of same radius.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create/update a ScGeom instance representing connexion between chained cylinders.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
If True, Cylinders nodes interact like spheres of radius 0.5*length, else one node has size length while the other has size 0. The difference is mainly the locus of rotation definition.
Enlarge both radii by this factor (if >1), to permit creation of distant interactions.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create/update geometry of collision between Facet and Polyhedra
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Incrementally compute L3Geom for contact between Facet and Sphere. Uses attributes of Ig2_Sphere_Sphere_L3Geom.
Selectively enable geometrical approximations (bitmask); add the values for approximations to be enabled.
1 | use previous transformation to transform velocities (which are known at mid-steps), instead of mid-step transformation computed as quaternion slerp at t=0.5. |
2 | do not take average (mid-step) normal when computing relative shear displacement, use previous value instead |
4 | do not re-normalize average (mid-step) normal, if used.… |
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Create interaction if spheres are not futher than distFactor *(r1+r2). If negative, zero normal deformation will be set to be the initial value (otherwise, the geometrical distance is the ‘’zero’’ one).
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
How often to renormalize trsf; if non-positive, never renormalized (simulation might be unstable)
Update object attributes from given dictionary
Create/update a ScGeom instance representing intersection of Facet and Sphere.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
The radius of the inscribed circle of the facet is decreased by the value of the sphere’s radius multipled by shrinkFactor. From the definition of contact point on the surface made of facets, the given surface is not continuous and becomes in effect surface covered with triangular tiles, with gap between the separate tiles equal to the sphere’s radius multiplied by 2×*shrinkFactor*. If zero, no shrinking is done.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create an interaction geometry ScGeom6D from Facet and Sphere, representing the Facet with a projected virtual sphere of same radius.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
The radius of the inscribed circle of the facet is decreased by the value of the sphere’s radius multipled by shrinkFactor. From the definition of contact point on the surface made of facets, the given surface is not continuous and becomes in effect surface covered with triangular tiles, with gap between the separate tiles equal to the sphere’s radius multiplied by 2×*shrinkFactor*. If zero, no shrinking is done.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create/update a GridCoGridCoGeom instance representing the geometry of a contact point between two GridConnection , including relative rotations.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create/update a GridNodeGeom6D instance representing the geometry of a contact point between two GridNode, including relative rotations.
Define relative velocity so that ratcheting is avoided. It applies for sphere-sphere contacts. It eventualy also apply for sphere-emulating interactions (i.e. convertible into the ScGeom type), if the virtual sphere’s motion is defined correctly (see e.g. Ig2_Sphere_ChainedCylinder_CylScGeom.
Short explanation of what we want to avoid :
Numerical ratcheting is best understood considering a small elastic cycle at a contact between two grains : assuming b1 is fixed, impose this displacement to b2 :
If the branch vector used to define the relative shear in rotation×branch is not constant (typically if it is defined from the vector center→contactPoint), then the shear displacement at the end of this cycle is not zero: rotations a and -a are multiplied by branches of different lengths.
It results in a finite contact force at the end of the cycle even though the positions and orientations are unchanged, in total contradiction with the elastic nature of the problem. It could also be seen as an inconsistent energy creation or loss. Given that DEM simulations tend to generate oscillations around equilibrium (damped mass-spring), it can have a significant impact on the evolution of the packings, resulting for instance in slow creep in iterations under constant load.
The solution adopted here to avoid ratcheting is as proposed by McNamara and co-workers. They analyzed the ratcheting problem in detail - even though they comment on the basis of a cycle that differs from the one shown above. One will find interesting discussions in e.g. [McNamara2008], even though solution it suggests is not fully applied here (equations of motion are not incorporating alpha, in contradiction with what is suggested by McNamara et al.).
Ordered list of types (as strings) this functor accepts.
Substract rotational creep from relative rotation. The rotational creep ScGeom6D::twistCreep is a quaternion and has to be updated inside a constitutive law, see for instance Law2_ScGeom6D_CohFrictPhys_CohesionMoment.
Return dictionary of attributes.
Enlarge both radii by this factor (if >1), to permit creation of distant interactions.
InteractionGeometry will be computed when interactionDetectionFactor*(rad1+rad2) > distance.
Note
This parameter is functionally coupled with Bo1_Sphere_Aabb::aabbEnlargeFactor, which will create larger bounding boxes and should be of the same value.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Precompute relative rotations. Turning this false can speed up simulations when rotations are not needed in constitutive laws (e.g. when spheres are compressed without cohesion and moment in early stage of a triaxial test), but is not foolproof. Change this value only if you know what you are doing.
Create/update geometry of collision between 2 Polyhedras
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create/update a ScGeom instance representing intersection of two Spheres.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Enlarge both radii by this factor (if >1), to permit creation of distant interactions.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create/update a ScGeom6D instance representing the geometry of a contact point between two Spheres, including relative rotations.
Ordered list of types (as strings) this functor accepts.
Substract rotational creep from relative rotation. The rotational creep ScGeom6D::twistCreep is a quaternion and has to be updated inside a constitutive law, see for instance Law2_ScGeom6D_CohFrictPhys_CohesionMoment.
Return dictionary of attributes.
Enlarge both radii by this factor (if >1), to permit creation of distant interactions.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Precompute relative rotations. Turning this false can speed up simulations when rotations are not needed in constitutive laws (e.g. when spheres are compressed without cohesion and moment in early stage of a triaxial test), but is not foolproof. Change this value only if you know what you are doing.
Create/update a ScGridCoGeom6D instance representing the geometry of a contact point between a GricConnection and a Sphere including relative rotations.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Enlarge both radii by this factor (if >1), to permit creation of distant interactions.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Functor for computing incrementally configuration of 2 Spheres stored in L3Geom; the configuration is positioned in global space by local origin (contact point) and rotation matrix
(orthonormal transformation matrix), and its degrees of freedom are local displacement
(in one normal and two shear directions); with Ig2_Sphere_Sphere_L6Geom and L6Geom, there is additionally
. The first row of
, i.e. local
-axis, is the contact normal noted
for brevity. Additionally, quasi-constant values of
(and
) are stored as shifted origins of
(and
); therefore, current value of displacement is always
.
Suppose two spheres with radii , positions
, velocities
, angular velocities
.
When there is not yet contact, it will be created if , where
is distFactor (sometimes also called
``interaction radius’‘). If
, then
will be initalized to
, otherwise to 0. In another words, contact will be created if spheres enlarged by
touch, and the ``equilibrium distance’’ (where
is zero) will be set to the current distance if
is positive, and to the geometrically-touching distance if negative.
Local axes (rows of ) are initially defined as follows:
If there has already been contact between the two spheres, it is updated to keep track of rigid motion of the contact (one that does not change mutual configuration of spheres) and mutual configuration changes. Rigid motion transforms local coordinate system and can be decomposed in rigid translation (affecting ), and rigid rotation (affecting
), which can be split in rotation
perpendicular to the normal and rotation
(``twist’‘) parallel with the normal:
Since velocities are known at previous midstep (), we consider mid-step normal
For the sake of numerical stability, is re-normalized after being computed, unless prohibited by approxMask. If approxMask has the appropriate bit set, the mid-normal is not compute, and we simply use
.
Rigid rotation parallel with the normal is
Branch vectors ,
(connecting
,
with
are computed depending on noRatch (see here).
Relative velocity at can be computed as
where is
without mean-field velocity gradient in periodic boundary conditions (see Cell.homoDeform). In the numerial implementation, the normal part of incident velocity is removed (since it is computed directly) with
.
Any vector expressed in global coordinates transforms during one timestep as
where the increments have the meaning of relative shear, rigid rotation normal to and rigid rotation parallel with
. Local coordinate system orientation, rotation matrix
, is updated by rows, i.e.
This matrix is re-normalized (unless prevented by approxMask) and mid-step transformation is computed using quaternion spherical interpolation as
Depending on approxMask, this computation can be avoided by approximating .
Finally, current displacement is evaluated as
For the normal component, non-incremental evaluation is preferred, giving
If this functor is called for L6Geom, local rotation is updated as
Selectively enable geometrical approximations (bitmask); add the values for approximations to be enabled.
1 | use previous transformation to transform velocities (which are known at mid-steps), instead of mid-step transformation computed as quaternion slerp at t=0.5. |
2 | do not take average (mid-step) normal when computing relative shear displacement, use previous value instead |
4 | do not re-normalize average (mid-step) normal, if used.… |
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Create interaction if spheres are not futher than distFactor *(r1+r2). If negative, zero normal deformation will be set to be the initial value (otherwise, the geometrical distance is the ‘’zero’’ one).
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
How often to renormalize trsf; if non-positive, never renormalized (simulation might be unstable)
Update object attributes from given dictionary
Incrementally compute L6Geom for contact of 2 spheres.
Selectively enable geometrical approximations (bitmask); add the values for approximations to be enabled.
1 | use previous transformation to transform velocities (which are known at mid-steps), instead of mid-step transformation computed as quaternion slerp at t=0.5. |
2 | do not take average (mid-step) normal when computing relative shear displacement, use previous value instead |
4 | do not re-normalize average (mid-step) normal, if used.… |
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Create interaction if spheres are not futher than distFactor *(r1+r2). If negative, zero normal deformation will be set to be the initial value (otherwise, the geometrical distance is the ‘’zero’’ one).
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
How often to renormalize trsf; if non-positive, never renormalized (simulation might be unstable)
Update object attributes from given dictionary
Create/update a ScGeom instance representing the geometry of a contact point between two Spheres s.
Define relative velocity so that ratcheting is avoided. It applies for sphere-sphere contacts. It eventualy also apply for sphere-emulating interactions (i.e. convertible into the ScGeom type), if the virtual sphere’s motion is defined correctly (see e.g. Ig2_Sphere_ChainedCylinder_CylScGeom.
Short explanation of what we want to avoid :
Numerical ratcheting is best understood considering a small elastic cycle at a contact between two grains : assuming b1 is fixed, impose this displacement to b2 :
If the branch vector used to define the relative shear in rotation×branch is not constant (typically if it is defined from the vector center→contactPoint), then the shear displacement at the end of this cycle is not zero: rotations a and -a are multiplied by branches of different lengths.
It results in a finite contact force at the end of the cycle even though the positions and orientations are unchanged, in total contradiction with the elastic nature of the problem. It could also be seen as an inconsistent energy creation or loss. Given that DEM simulations tend to generate oscillations around equilibrium (damped mass-spring), it can have a significant impact on the evolution of the packings, resulting for instance in slow creep in iterations under constant load.
The solution adopted here to avoid ratcheting is as proposed by McNamara and co-workers. They analyzed the ratcheting problem in detail - even though they comment on the basis of a cycle that differs from the one shown above. One will find interesting discussions in e.g. [McNamara2008], even though solution it suggests is not fully applied here (equations of motion are not incorporating alpha, in contradiction with what is suggested by McNamara et al.).
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Enlarge both radii by this factor (if >1), to permit creation of distant interactions.
InteractionGeometry will be computed when interactionDetectionFactor*(rad1+rad2) > distance.
Note
This parameter is functionally coupled with Bo1_Sphere_Aabb::aabbEnlargeFactor, which will create larger bounding boxes and should be of the same value.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create/update a ScGeom6D instance representing the geometry of a contact point between two Spheres, including relative rotations.
Define relative velocity so that ratcheting is avoided. It applies for sphere-sphere contacts. It eventualy also apply for sphere-emulating interactions (i.e. convertible into the ScGeom type), if the virtual sphere’s motion is defined correctly (see e.g. Ig2_Sphere_ChainedCylinder_CylScGeom.
Short explanation of what we want to avoid :
Numerical ratcheting is best understood considering a small elastic cycle at a contact between two grains : assuming b1 is fixed, impose this displacement to b2 :
If the branch vector used to define the relative shear in rotation×branch is not constant (typically if it is defined from the vector center→contactPoint), then the shear displacement at the end of this cycle is not zero: rotations a and -a are multiplied by branches of different lengths.
It results in a finite contact force at the end of the cycle even though the positions and orientations are unchanged, in total contradiction with the elastic nature of the problem. It could also be seen as an inconsistent energy creation or loss. Given that DEM simulations tend to generate oscillations around equilibrium (damped mass-spring), it can have a significant impact on the evolution of the packings, resulting for instance in slow creep in iterations under constant load.
The solution adopted here to avoid ratcheting is as proposed by McNamara and co-workers. They analyzed the ratcheting problem in detail - even though they comment on the basis of a cycle that differs from the one shown above. One will find interesting discussions in e.g. [McNamara2008], even though solution it suggests is not fully applied here (equations of motion are not incorporating alpha, in contradiction with what is suggested by McNamara et al.).
Ordered list of types (as strings) this functor accepts.
Substract rotational creep from relative rotation. The rotational creep ScGeom6D::twistCreep is a quaternion and has to be updated inside a constitutive law, see for instance Law2_ScGeom6D_CohFrictPhys_CohesionMoment.
Return dictionary of attributes.
Enlarge both radii by this factor (if >1), to permit creation of distant interactions.
InteractionGeometry will be computed when interactionDetectionFactor*(rad1+rad2) > distance.
Note
This parameter is functionally coupled with Bo1_Sphere_Aabb::aabbEnlargeFactor, which will create larger bounding boxes and should be of the same value.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Precompute relative rotations. Turning this false can speed up simulations when rotations are not needed in constitutive laws (e.g. when spheres are compressed without cohesion and moment in early stage of a triaxial test), but is not foolproof. Change this value only if you know what you are doing.
Create/update geometry of collision between 2 tetrahedra (TTetraGeom instance)
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
EXPERIMANTAL. Create/update geometry of collision between 2 tetrahedra (TTetraSimpleGeom instance)
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create/update geometry of collision between Wall and Polyhedra
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Incrementally compute L3Geom for contact between Wall and Sphere. Uses attributes of Ig2_Sphere_Sphere_L3Geom.
Selectively enable geometrical approximations (bitmask); add the values for approximations to be enabled.
1 | use previous transformation to transform velocities (which are known at mid-steps), instead of mid-step transformation computed as quaternion slerp at t=0.5. |
2 | do not take average (mid-step) normal when computing relative shear displacement, use previous value instead |
4 | do not re-normalize average (mid-step) normal, if used.… |
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Create interaction if spheres are not futher than distFactor *(r1+r2). If negative, zero normal deformation will be set to be the initial value (otherwise, the geometrical distance is the ‘’zero’’ one).
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
How often to renormalize trsf; if non-positive, never renormalized (simulation might be unstable)
Update object attributes from given dictionary
Create/update a ScGeom instance representing intersection of Wall and Sphere.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Avoid granular ratcheting
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Dispatcher calling functors based on received argument type(s).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Return functor that would be dispatched for given argument(s); None if no dispatch; ambiguous dispatch throws.
Return dictionary with contents of the dispatch matrix.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Functors associated with this dispatcher.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Functor for creating/updating Interaction::phys objects.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Generates cohesive-frictional interactions with moments. Used in the contact law Law2_ScGeom6D_InelastCohFrictPhys_CohesionMoment.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
The RelationShips for using Law2_ScGeom6D_NormalInelasticityPhys_NormalInelasticity
In these RelationShips all the attributes of the interactions (which are of NormalInelasticityPhys type) are computed.
Warning
as in the others Ip2 functors, most of the attributes are computed only once, when the interaction is new.
Ordered list of types (as strings) this functor accepts.
Parameter for computing the torque-stifness : T-stifness = betaR * Rmoy^2
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Generates bubble interactions.Used in the contact law Law2_ScGeom_BubblePhys_Bubble.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Generates cohesive-frictional interactions with moments. Used in the contact law Law2_ScGeom6D_CohFrictPhys_CohesionMoment. The normal/shear stiffness and friction definitions are the same as in Ip2_FrictMat_FrictMat_FrictPhys, check the documentation there for details.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
If true, assign cohesion to all existing contacts in current time-step. The flag is turned false automatically, so that assignment is done in the current timestep only.
If true, assign cohesion at all new contacts. If false, only existing contacts can be cohesive (also see Ip2_CohFrictMat_CohFrictMat_CohFrictPhys::setCohesionNow), and new contacts are only frictional.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Convert 2 CpmMat instances to CpmPhys with corresponding parameters. Uses simple (arithmetic) averages if material are different. Simple copy of parameters is performed if the material is shared between both particles. See cpm-model for detals.
Ordered list of types (as strings) this functor accepts.
Should new contacts be cohesive? They will before this iter#, they will not be afterwards. If 0, they will never be. If negative, they will always be created as cohesive (10 by default).
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create a NormPhys from two ElastMats. TODO. EXPERIMENTAL
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create a NormShearPhys from two ElastMats. TODO. EXPERIMENTAL
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Convert CpmMat instance and FrictMat instance to FrictPhys with corresponding parameters (young, poisson, frictionAngle). Uses simple (arithmetic) averages if material parameters are different.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
RelationShips to use with Law2_ScGeom_CapillaryPhys_Capillarity
In these RelationShips all the interaction attributes are computed.
Warning
as in the others Ip2 functors, most of the attributes are computed only once, when the interaction is new.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create a FrictPhys from two FrictMats. The compliance of one sphere under point load is defined here as , with
the stiffness of the sphere and
its diameter. The compliance of the contact itself will be the sum of compliances from each sphere, i.e.
in the general case, or
in the special case of equal sizes and equal stiffness. Note that summing compliances corresponds to an harmonic average of stiffnesss (as in e.g. [Scholtes2009a]), which is how kn is actually computed in the Ip2_FrictMat_FrictMat_FrictPhys functor:
, with
.
The shear stiffness ks of one sphere is defined via the material parameter ElastMat::poisson, as ks=poisson*kn, and the resulting shear stiffness of the interaction will be also an harmonic average.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Instance of MatchMaker determining how to compute interaction’s friction angle. If None, minimum value is used.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
RelationShips to use with Law2_ScGeom_CapillaryPhys_Capillarity
In these RelationShips all the interaction attributes are computed.
Warning
as in the others Ip2 functors, most of the attributes are computed only once, when the interaction is new.
Ordered list of types (as strings) this functor accepts.
Normal viscous damping coefficient .
Shear viscous damping coefficient .
Return dictionary of attributes.
Normal coefficient of restitution .
Shear coefficient of restitution .
Coefficient to determine the plastic bending moment
Surface energy parameter [J/m^2] per each unit contact surface, to derive DMT formulation from HM
Rotational stiffness for moment contact law
Torsional stiffness for moment contact law
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Calculate some physical parameters needed to obtain the normal and shear stiffnesses according to the Hertz-Mindlin’s formulation (as implemented in PFC).
Viscous parameters can be specified either using coefficients of restitution (,
) or viscous damping coefficient (
,
). The following rules apply:
#. If the
(
) coefficient is given, it is assigned to MindlinPhys.betan (MindlinPhys.betas) directly.
#. If
is given, MindlinPhys.betan is computed using
. The same applies to
, MindlinPhys.betas.
#. It is an error (exception) to specify both
and
(
and
).
#. If neither
nor
is given, zero value for MindlinPhys.betan is used; there will be no viscous effects.
#.If neither
nor
is given, the value of MindlinPhys.betan is used for MindlinPhys.betas as well.
The ,
,
,
are MatchMaker objects; they can be constructed from float values to always return constant value.
See scripts/test/shots.py for an example of specifying based on combination of parameters.
Ordered list of types (as strings) this functor accepts.
Normal viscous damping coefficient .
Shear viscous damping coefficient .
Return dictionary of attributes.
Normal coefficient of restitution .
Shear coefficient of restitution .
Coefficient to determine the plastic bending moment
Instance of MatchMaker determining how to compute the friction angle of an interaction. If None, minimum value is used.
Surface energy parameter [J/m^2] per each unit contact surface, to derive DMT formulation from HM
Rotational stiffness for moment contact law
Torsional stiffness for moment contact law
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Create a FrictPhys from two FrictMats. The compliance of one sphere under symetric point loads is defined here as 1/(E.r), with E the stiffness of the sphere and r its radius, and corresponds to a compliance 1/(2.E.r)=1/(E.D) from each contact point. The compliance of the contact itself will be the sum of compliances from each sphere, i.e. 1/(E.D1)+1/(E.D2) in the general case, or 1/(E.r) in the special case of equal sizes. Note that summing compliances corresponds to an harmonic average of stiffnesss, which is how kn is actually computed in the Ip2_FrictMat_FrictMat_FrictPhys functor.
The shear stiffness ks of one sphere is defined via the material parameter ElastMat::poisson, as ks=poisson*kn, and the resulting shear stiffness of the interaction will be also an harmonic average.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Instance of MatchMaker determining how to compute interaction’s friction angle. If None, minimum value is used.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Converts 2 JCFpmMat instances to one JCFpmPhys instance, with corresponding parameters.
Ordered list of types (as strings) this functor accepts.
should new contacts be cohesive? They will before this iter, they won’t afterward.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Convert 2 instances of LudingMat to LudingPhys using the rule of consecutive connection.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Convert 2 instances of ViscElMat to ViscElPhys using the rule of consecutive connection.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Converts 2 WireMat instances to WirePhys with corresponding parameters.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Iteration to create the link.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Dispatcher calling functors based on received argument type(s).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Return functor that would be dispatched for given argument(s); None if no dispatch; ambiguous dispatch throws.
Return dictionary with contents of the dispatch matrix.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Functors associated with this dispatcher.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Functor for applying constitutive laws on interactions.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Law for linear compression, and Mohr-Coulomb plasticity surface without cohesion.
This law implements the classical linear elastic-plastic law from [CundallStrack1979] (see also [Pfc3dManual30]). The normal force is (with the convention of positive tensile forces) . The shear force is
, the plasticity condition defines the maximum value of the shear force :
, with
the friction angle.
Note
This law is well tested in the context of triaxial simulation, and has been used for a number of published results (see e.g. [Scholtes2009b] and other papers from the same authors). It is generalised by Law2_ScGeom6D_CohFrictPhys_CohesionMoment, which adds cohesion and moments at contact.
If true, use bending/twisting moments at all contacts. If false, compute moments only for cohesive contacts.
Ordered list of types (as strings) this functor accepts.
creep viscosity [Pa.s/m]. probably should be moved to Ip2_CohFrictMat_CohFrictMat_CohFrictPhys...
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
activate creep on the shear force, using CohesiveFrictionalContactLaw::creep_viscosity.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
activate creep on the twisting moment, using CohesiveFrictionalContactLaw::creep_viscosity.
Update object attributes from given dictionary
use the incremental formulation to compute bending and twisting moments. Creep on the twisting moment is not included in such a case.
This law generalises Law2_CylScGeom_FrictPhys_CundallStrack by adding cohesion and moments at contact.
If true, use bending/twisting moments at all contacts. If false, compute moments only for cohesive contacts.
Ordered list of types (as strings) this functor accepts.
creep viscosity [Pa.s/m]. probably should be moved to Ip2_CohFrictMat_CohFrictMat_CohFrictPhys...
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
activate creep on the shear force, using CohesiveFrictionalContactLaw::creep_viscosity.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
activate creep on the twisting moment, using CohesiveFrictionalContactLaw::creep_viscosity.
Update object attributes from given dictionary
use the incremental formulation to compute bending and twisting moments. Creep on the twisting moment is not included in such a case.
Law for linear compression, and Mohr-Coulomb plasticity surface without cohesion.
This law implements the classical linear elastic-plastic law from [CundallStrack1979] (see also [Pfc3dManual30]). The normal force is (with the convention of positive tensile forces) . The shear force is
, the plasticity condition defines the maximum value of the shear force :
, with
the friction angle.
Note
This law uses ScGeom.
Note
This law is well tested in the context of triaxial simulation, and has been used for a number of published results (see e.g. [Scholtes2009b] and other papers from the same authors). It is generalised by Law2_ScGeom6D_CohFrictPhys_CohesionMoment, which adds cohesion and moments at contact.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Frictional elastic contact law between two gridConnection . See Law2_ScGeom_FrictPhys_CundallStrack for more details.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Compute and return the total elastic energy in all “FrictPhys” contacts
Initialize cummulated plastic dissipation to a value (0 by default).
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Total energy dissipated in plastic slips at all FrictPhys contacts. Computed only if Law2_ScGeom_FrictPhys_CundallStrack::traceEnergy is true.
If true, compute branch vectors from radii (faster), else use contactPoint-position. Turning this flag true is safe for sphere-sphere contacts and a few other specific cases. It will give wrong values of torques on facets or boxes.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Define the total energy dissipated in plastic slips at all contacts. This will trace only plastic energy in this law, see O.trackEnergy for a more complete energies tracing
Update object attributes from given dictionary
Basic law for testing L3Geom; it bears no cohesion (unless noBreak is True), and plastic slip obeys the Mohr-Coulomb criterion (unless noSlip is True).
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Do not break contacts when particles separate.
No plastic slipping.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Basic law for testing L6Geom – linear in both normal and shear sense, without slip or breakage.
Ordered list of types (as strings) this functor accepts.
Characteristic length with the meaning of the stiffness ratios bending/shear and torsion/normal.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Do not break contacts when particles separate.
No plastic slipping.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Law for linear traction-compression-bending-twisting, with cohesion+friction and Mohr-Coulomb plasticity surface. This law adds adhesion and moments to Law2_ScGeom_FrictPhys_CundallStrack.
The normal force is (with the convention of positive tensile forces) , with
the normal adhesion. The shear force is
, the plasticity condition defines the maximum value of the shear force, by default
, with
the friction angle and
the shear adhesion. If CohFrictPhys::cohesionDisableFriction is True, friction is ignored as long as adhesion is active, and the maximum shear force is only
.
If the maximum tensile or maximum shear force is reached and CohFrictPhys::fragile =True (default), the cohesive link is broken, and are set back to zero. If a tensile force is present, the contact is lost, else the shear strength is
. If CohFrictPhys::fragile =False (in course of implementation), the behaviour is perfectly plastic, and the shear strength is kept constant.
If Law2_ScGeom6D_CohFrictPhys_CohesionMoment::momentRotationLaw =True, bending and twisting moments are computed using a linear law with moduli respectively and
(the two values are the same currently), so that the moments are :
and
, with
the relative rotations between interacting bodies. The maximum value of moments can be defined and takes the form of rolling friction. Cohesive -type moment may also be included in the future.
Creep at contact is implemented in this law, as defined in [Hassan2010]. If activated, there is a viscous behaviour of the shear and twisting components, and the evolution of the elastic parts of shear displacement and relative twist is given by and
.
If true, use bending/twisting moments at all contacts. If false, compute moments only for cohesive contacts.
Ordered list of types (as strings) this functor accepts.
creep viscosity [Pa.s/m]. probably should be moved to Ip2_CohFrictMat_CohFrictMat_CohFrictPhys.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Compute normal elastic energy.
Compute shear elastic energy.
activate creep on the shear force, using CohesiveFrictionalContactLaw::creep_viscosity.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
activate creep on the twisting moment, using CohesiveFrictionalContactLaw::creep_viscosity.
Update object attributes from given dictionary
use the incremental formulation to compute bending and twisting moments. Creep on the twisting moment is not included in such a case.
Law for linear traction-compression-bending-twisting, with cohesion+friction and Mohr-Coulomb plasticity surface. This law adds adhesion and moments to Law2_ScGeom_FrictPhys_CundallStrack.
The normal force is (with the convention of positive tensile forces) , with
the normal adhesion. The shear force is
, the plasticity condition defines the maximum value of the shear force, by default
, with
the friction angle and
the shear adhesion. If CohFrictPhys::cohesionDisableFriction is True, friction is ignored as long as adhesion is active, and the maximum shear force is only
.
If the maximum tensile or maximum shear force is reached and CohFrictPhys::fragile =True (default), the cohesive link is broken, and are set back to zero. If a tensile force is present, the contact is lost, else the shear strength is
. If CohFrictPhys::fragile =False (in course of implementation), the behaviour is perfectly plastic, and the shear strength is kept constant.
If Law2_ScGeom6D_CohFrictPhys_CohesionMoment::momentRotationLaw =True, bending and twisting moments are computed using a linear law with moduli respectively and
(the two values are the same currently), so that the moments are :
and
, with
the relative rotations between interacting bodies. The maximum value of moments can be defined and takes the form of rolling friction. Cohesive -type moment may also be included in the future.
Creep at contact is implemented in this law, as defined in [Hassan2010]. If activated, there is a viscous behaviour of the shear and twisting components, and the evolution of the elastic parts of shear displacement and relative twist is given by and
.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Compute normal elastic energy.
Compute shear elastic energy.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
use the incremental formulation to compute bending and twisting moments. Creep on the twisting moment is not included in such a case.
Contact law used to simulate granular filler in rock joints [Duriez2009a], [Duriez2011]. It includes possibility of cohesion, moment transfer and inelastic compression behaviour (to reproduce the normal inelasticity observed for rock joints, for the latter).
The moment transfer relation corresponds to the adaptation of the work of Plassiard & Belheine (see in [DeghmReport2006] for example), which was realized by J. Kozicki, and is now coded in ScGeom6D.
As others LawFunctor, it uses pre-computed data of the interactions (rigidities, friction angles -with their tan()-, orientations of the interactions); this work is done here in Ip2_2xNormalInelasticMat_NormalInelasticityPhys.
To use this you should also use NormalInelasticMat as material type of the bodies.
The effects of this law are illustrated in examples/normalInelasticityTest.py
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
boolean, true=> the part of the contact torque (caused by relative rotations, which is computed only if momentRotationLaw..) is not limited by a plastic threshold
boolean, true=> computation of a torque (against relative rotation) exchanged between particles
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Constitutive law for Bubble model.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Constitutive law for the cpm-model.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Compute and return the total elastic energy in all “CpmPhys” contacts
Strain at which softening in compression starts (non-negative to deactivate)
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
damage after which the contact disappears (<1), since omega reaches 1 only for strain →+∞
Relative rigidity of the softening branch in compression (0=perfect elastic-plastic, <0 softening, >0 hardening)
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
horizontal scaling of the ellipse (shifts on the +x axis as interactions with +y are given)
scaling in the logarithmic yield surface (should be <1 for realistic results; >=0 for meaningful results)
Return radius of yield surface for given material and state parameters; uses attributes of the current instance (yieldSurfType etc), change them before calling if you need that.
yield function: 0: mohr-coulomb (original); 1: parabolic; 2: logarithmic, 3: log+lin_tension, 4: elliptic, 5: elliptic+log
Law for linear compression, and Mohr-Coulomb plasticity surface without cohesion.
This law implements the classical linear elastic-plastic law from [CundallStrack1979] (see also [Pfc3dManual30]). The normal force is (with the convention of positive tensile forces) . The shear force is
, the plasticity condition defines the maximum value of the shear force :
, with
the friction angle.
This law is well tested in the context of triaxial simulation, and has been used for a number of published results (see e.g. [Scholtes2009b] and other papers from the same authors). It is generalised by Law2_ScGeom6D_CohFrictPhys_CohesionMoment, which adds cohesion and moments at contact.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Compute and return the total elastic energy in all “FrictPhys” contacts
Initialize cummulated plastic dissipation to a value (0 by default).
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Total energy dissipated in plastic slips at all FrictPhys contacts. Computed only if Law2_ScGeom_FrictPhys_CundallStrack::traceEnergy is true.
If true, compute branch vectors from radii (faster), else use contactPoint-position. Turning this flag true is safe for sphere-sphere contacts and a few other specific cases. It will give wrong values of torques on facets or boxes.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Define the total energy dissipated in plastic slips at all contacts. This will trace only plastic energy in this law, see O.trackEnergy for a more complete energies tracing
Update object attributes from given dictionary
Interaction law for cohesive frictional material, e.g. rock, possibly presenting joint surfaces. Joint surfaces can be defined in a preprocessing phase through .stl meshes (see ref for details of the procedure), and can be mechanically described with a smooth contact logic [Ivars2011] (implemented in Yade in [Scholtes2012]).
string specifying the name of saved file ‘cracks___.txt’, when recordCracks is true.
Ordered list of types (as strings) this functor accepts.
if true (and if recordCracks), data are appended to an existing ‘cracksKey’ text file; otherwise its content is reset.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
if true a text file cracksKey.txt (see Key) will be created. It contains : apparition iteration, position, type (tensile or shear), cross section and contact normal.
if true, interactions of particles belonging to joint surface (JCFpmPhys.isOnJoint) are handled according to a smooth contact logic [Ivars2011], [Scholtes2012].
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Linear viscoelastic model operating on ScGeom and LudingPhys.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Constitutive law for the Hertz formulation (using MindlinPhys.kno) and linear beahvior in shear (using MindlinPhys.kso for stiffness and FrictPhys.tangensOfFrictionAngle).
Note
No viscosity or damping. If you need those, look at Law2_ScGeom_MindlinPhys_Mindlin, which also includes non-linear Mindlin shear.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Shear force nonlinearity (the value determines how many features of the non-linearity are taken in account). 1: ks as in HM 2: shearElastic increment computed as in HM 3. granular ratcheting disabled.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Constitutive law for the Hertz-Mindlin formulation. It includes non linear elasticity in the normal direction as predicted by Hertz for two non-conforming elastic contact bodies. In the shear direction, instead, it reseambles the simplified case without slip discussed in Mindlin’s paper, where a linear relationship between shear force and tangential displacement is provided. Finally, the Mohr-Coulomb criterion is employed to established the maximum friction force which can be developed at the contact. Moreover, it is also possible to include the effect of linear viscous damping through the definition of the parameters and
.
Ordered list of types (as strings) this functor accepts.
bool to calculate energy terms (shear potential energy, dissipation of energy due to friction and dissipation of energy due to normal and tangential damping)
Compute total number of adhesive contacts.
Return dictionary of attributes.
Energy dissipation due to sliding
bool to include the adhesion force following the DMT formulation. If true, also the normal elastic energy takes into account the adhesion effect.
bool to consider rolling resistance (if Ip2_FrictMat_FrictMat_MindlinPhys::eta is 0.0, no plastic condition is applied.)
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Energy dissipated by normal damping
Compute normal elastic potential energy. It handles the DMT formulation if Law2_ScGeom_MindlinPhys_Mindlin::includeAdhesion is set to true.
bool to avoid granular ratcheting
Return the ratio between the number of contacts sliding to the total number at a given time.
Energy dissipated by tangential damping
Shear elastic potential energy
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Hertz-Mindlin contact law with partial slip solution, as described in [Thornton1991].
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Linear viscoelastic model operating on ScGeom and ViscElPhys. The model is mostly based on the paper for For details see Pournin [Pournin2001] .
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Law similar to Law2_ScGeom_FrictPhys_CundallStrack with the addition of shear creep at contacts.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Compute and return the total elastic energy in all “FrictPhys” contacts
Initialize cummulated plastic dissipation to a value (0 by default).
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Total energy dissipated in plastic slips at all FrictPhys contacts. Computed only if Law2_ScGeom_FrictPhys_CundallStrack::traceEnergy is true.
If true, compute branch vectors from radii (faster), else use contactPoint-position. Turning this flag true is safe for sphere-sphere contacts and a few other specific cases. It will give wrong values of torques on facets or boxes.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Define the total energy dissipated in plastic slips at all contacts. This will trace only plastic energy in this law, see O.trackEnergy for a more complete energies tracing
Update object attributes from given dictionary
Constitutive law for the wire model.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Iteration to create the link.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Law between a cohesive frictional GridConnection and a cohesive frictional Sphere. Almost the same than Law2_ScGeom6D_CohFrictPhys_CohesionMoment, but THE ROTATIONAL MOMENTS ARE NOT COMPUTED.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Law between a frictional GridConnection and a frictional Sphere. Almost the same than Law2_ScGeom_FrictPhys_CundallStrack, but the force is divided and applied on the two GridNodes only.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Keep interactions even if particles go away from each other (only in case another constitutive law is in the scene, e.g. Law2_ScGeom_CapillaryPhys_Capillarity)
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
EXPERIMENTAL. TODO
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Calculate physical response of 2 vector in interaction, based on penetration configuration given by PolyhedraGeom.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Compute and return the total elastic energy in all “FrictPhys” contacts
Initialize cummulated plastic dissipation to a value (0 by default).
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Total energy dissipated in plastic slips at all FrictPhys contacts. Computed only if Law2_ScGeom_FrictPhys_CundallStrack::traceEnergy is true.
Shear force from last step
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Define the total energy dissipated in plastic slips at all contacts. This will trace only plastic energy in this law, see O.trackEnergy for a more complete energies tracing
Update object attributes from given dictionary
Dispatcher calling functors based on received argument type(s).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Return functor that would be dispatched for given argument(s); None if no dispatch; ambiguous dispatch throws.
Return dictionary with contents of the dispatch matrix.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Functors associated with this dispatcher.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Abstract callback object which will be called for every (real) Interaction after the interaction has been processed by InteractionLoop.
At the beginning of the interaction loop, stepInit is called, initializing the object; it returns either NULL (to deactivate the callback during this time step) or pointer to function, which will then be passed (1) pointer to the callback object itself and (2) pointer to Interaction.
Note
(NOT YET DONE) This functionality is accessible from python by passing 4th argument to InteractionLoop constructor, or by appending the callback object to InteractionLoop::callbacks.
Return dictionary of attributes.
Update object attributes from given dictionary
Callback summing magnitudes of forces over all interactions. IPhys of interactions must derive from NormShearPhys (responsability fo the user).
Return dictionary of attributes.
Update object attributes from given dictionary
Base class for scene generators, preprocessors.
Return dictionary of attributes.
Generate scene, save to given file
Generate scene, save to temporary file and load immediately
Update object attributes from given dictionary
This preprocessor is a variant of TriaxialTest, including the model of capillary forces developed as part of the PhD of Luc Scholtès. See the documentation of Law2_ScGeom_CapillaryPhys_Capillarity or the main page https://yade-dem.org/wiki/CapillaryTriaxialTest, for more details.
Results obtained with this preprocessor were reported for instance in ‘Scholtes et al. Micromechanics of granular materials with capillary effects. International Journal of Engineering Science 2009,(47)1, 64-75.’
A code that is added to output filenames.
Normalized standard deviation of generated sizes.
Value of unbalanced force for which the system is considered stable. Used in conditionals to switch between loading stages.
Do we just want to generate a stable packing under isotropic pressure (false) or do we want the triaxial loading to start automatically right after compaction stage (true)?
freeze the simulation when conditions are reached (don’t activate this if you want to be able to run/stop from Qt GUI)
auto adjust the isotropic stress state from TriaxialTest::sigmaIsoCompaction to TriaxialTest::sigmaLateralConfinement if they have different values. See docs for TriaxialCompressionEngine::autoUnload
FIXME : what is that?
Defines how overlapping bridges affect the capillary forces (see CapillaryTriaxialTest::fusionDetection). If binary=true, the force is null as soon as there is an overlap detected, if not, the force is divided by the number of overlaps.
Friction angle [°] of boundaries contacts.
Ratio of shear vs. normal contact stiffness for boxes.
Use boxes for boundaries (recommended).
Stiffness of boxes.
Define succion in the packing [Pa]. This is the value used in the capillary model.
Friction angle [°] of spheres during compaction (different values result in different porosities)]. This value is overridden by TriaxialTest::sphereFrictionDeg before triaxial testing.
Coefficient of Cundal-Non-Viscous damping (applied on on the 3 components of forces)
Coefficient of Cundal-Non-Viscous damping (applied on on the 3 components of torques)
Max time-step. Used as initial value if defined. Latter adjusted by the time stepper.
density of spheres
Return dictionary of attributes.
Use facets for boundaries (not tested)
max multiplier of diameters during internal compaction (secondary precise adjustment)
string that contains some subset (max. 2) of {‘x’,’y’,’z’} ; contains axes will have box dimension hardcoded, even if box is scaled as mean_radius is prescribed: scaling will be applied on the rest.
flag to choose an isotropic compaction until a fixed porosity choosing a same translation speed for the six walls
FIXME : what is that?
test overlaps between liquid bridges on modify forces if overlaps exist
Generate scene, save to given file
File with positions and sizes of spheres.
flag for choosing between moving boundaries or increasing particles sizes during the compaction stage.
Generate scene, save to temporary file and load immediately
Lower corner of the box.
max multiplier of diameters during internal compaction (initial fast increase)
max velocity of boundaries. Usually useless, but can help stabilizing the system in some cases.
Do not create any files during run (.xml, .spheres, wall stress records)
Number of generated spheres.
interval between size changes when growing spheres.
Mean radius. If negative (default), autocomputed to as a function of box size and TriaxialTest::numberOfGrains
interval between file outputs
Confining stress during isotropic compaction.
Lateral stress during triaxial loading. An isotropic unloading is performed if the value is not equal to CapillaryTriaxialTest::SigmaIsoCompaction.
Friction angle [°] of spheres assigned just before triaxial testing.
Ratio of shear vs. normal contact stiffness for spheres.
Stiffness of spheres.
Strain rate in triaxial loading.
thickness of boundaries. It is arbitrary and should have no effect
interval for outputing general information on the simulation (stress,unbalanced force,...)
interval for GlobalStiffnessTimeStepper
Update object attributes from given dictionary
Upper corner of the box.
Make boundaries larger than the packing to make sure spheres don’t go out during deformation.
interval for updating the stiffness of sample/boundaries contacts
Use walls for boundaries (not tested)
activate capillary model
This preprocessor is a variant of TriaxialTest using the cohesive-frictional contact law with moments. It sets up a scene for cohesive triaxial tests. See full documentation at http://yade-dem.org/wiki/TriaxialTest.
Cohesion is initially 0 by default. The suggested usage is to define cohesion values in a second step, after isotropic compaction : define shear and normal cohesions in Ip2_CohFrictMat_CohFrictMat_CohFrictPhys, then turn Ip2_CohFrictMat_CohFrictMat_CohFrictPhys::setCohesionNow true to assign them at each contact at next iteration.
A code that is added to output filenames.
Value of unbalanced force for which the system is considered stable. Used in conditionals to switch between loading stages.
Do we just want to generate a stable packing under isotropic pressure (false) or do we want the triaxial loading to start automatically right after compaction stage (true)?
freeze the simulation when conditions are reached (don’t activate this if you want to be able to run/stop from Qt GUI)
auto adjust the isotropic stress state from TriaxialTest::sigmaIsoCompaction to TriaxialTest::sigmaLateralConfinement if they have different values. See docs for TriaxialCompressionEngine::autoUnload
FIXME : what is that?
Friction angle [°] of boundaries contacts.
Ratio of shear vs. normal contact stiffness for boxes.
Use boxes for boundaries (recommended).
Stiffness of boxes.
Friction angle [°] of spheres during compaction (different values result in different porosities)]. This value is overridden by TriaxialTest::sphereFrictionDeg before triaxial testing.
Coefficient of Cundal-Non-Viscous damping (applied on on the 3 components of forces)
Coefficient of Cundal-Non-Viscous damping (applied on on the 3 components of torques)
Max time-step. Used as initial value if defined. Latter adjusted by the time stepper.
density of spheres
Return dictionary of attributes.
Use facets for boundaries (not tested)
max multiplier of diameters during internal compaction (secondary precise adjustment)
string that contains some subset (max. 2) of {‘x’,’y’,’z’} ; contains axes will have box dimension hardcoded, even if box is scaled as mean_radius is prescribed: scaling will be applied on the rest.
flag to choose an isotropic compaction until a fixed porosity choosing a same translation speed for the six walls
FIXME : what is that?
Generate scene, save to given file
File with positions and sizes of spheres.
flag for choosing between moving boundaries or increasing particles sizes during the compaction stage.
Generate scene, save to temporary file and load immediately
Lower corner of the box.
max multiplier of diameters during internal compaction (initial fast increase)
max velocity of boundaries. Usually useless, but can help stabilizing the system in some cases.
Do not create any files during run (.xml, .spheres, wall stress records)
Material parameter used to define contact strength in tension.
Number of generated spheres.
interval between size changes when growing spheres.
Normalized standard deviation of generated sizes.
Mean radius. If negative (default), autocomputed to as a function of box size and TriaxialTest::numberOfGrains
interval between file outputs
create cohesionless (False) or cohesive (True) interactions for new contacts.
Material parameter used to define shear strength of contacts.
Confining stress during isotropic compaction.
Lateral stress during triaxial loading. An isotropic unloading is performed if the value is not equal to TriaxialTest::sigmaIsoCompaction.
Friction angle [°] of spheres assigned just before triaxial testing.
Ratio of shear vs. normal contact stiffness for spheres.
Stiffness of spheres.
Strain rate in triaxial loading.
thickness of boundaries. It is arbitrary and should have no effect
interval for GlobalStiffnessTimeStepper
Update object attributes from given dictionary
Upper corner of the box.
Make boundaries larger than the packing to make sure spheres don’t go out during deformation.
interval for updating the stiffness of sample/boundaries contacts
Use walls for boundaries (not tested)
Preprocessor for creating a numerical model of a simple shear box.
Launching this preprocessor will carry out an oedometric compression, until a value of normal stress equal to 2 MPa (and stable). But with others Engines KinemCNDEngine, KinemCNSEngine and KinemCNLEngine, respectively constant normal displacement, constant normal rigidity and constant normal stress paths can be carried out for such simple shear boxes.
NB about micro-parameters : their default values correspond to those used in [Duriez2009a] and [Duriez2011] to simulate infilled rock joints.
value of ElastMat::poisson for the spheres [-]
value of ElastMat::young for the boxes []
density of the spheres []
Return dictionary of attributes.
Generate scene, save to given file
depending on this, GravityEngine is added or not to the scene to take into account the weight of particles
vector corresponding to used gravity (if gravApplied) []
initial height (along y-axis) of the shear box []
initial length (along x-axis) of the shear box []
Generate scene, save to temporary file and load immediately
value of ElastMat::poisson for the spheres [] (the necessary conversion in
is done automatically)
value of ElastMat::poisson for the spheres [-]
value of ElastMat::young for the spheres []
thickness of the boxes constituting the shear box []
value of TimeStepper::timeStepUpdateInterval for the TimeStepper used here
Update object attributes from given dictionary
initial width (along z-axis) of the shear box []
Create a scene for triaxal test.
Yade includes tools to simulate triaxial tests on particles assemblies. This pre-processor (and variants like e.g. CapillaryTriaxialTest) illustrate how to use them. It generates a scene which will - by default - go through the following steps :
The default loading path corresponds to a constant lateral stress (sigmaLateralConfinement) in 2 directions and constant strain rate on the third direction. This default loading path is performed when the flag autoCompressionActivation it True, otherwise the simulation stops after isotropic compression.
Different loading paths might be performed. In order to define them, the user can modify the flags found in engine TriaxialStressController at any point in the simulation (in c++). If TriaxialStressController.wall_X_activated is true boundary X is moved automatically to maintain the defined stress level sigmaN (see axis conventions below). If false the boundary is not controlled by the engine at all. In that case the user is free to prescribe fixed position, constant velocity, or more complex conditions.
Note
Axis conventions. Boundaries perpendicular to the x axis are called “left” and “right”, y corresponds to “top” and “bottom”, and axis z to “front” and “back”. In the default loading path, strain rate is assigned along y, and constant stresses are assigned on x and z.
Note
TriaxialStressController::ComputeUnbalancedForce returns a value that can be useful for evaluating the stability of the packing. It is defined as (mean force on particles)/(mean contact force), so that it tends to 0 in a stable packing. This parameter is checked by TriaxialCompressionEngine to switch from one stage of the simulation to the next one (e.g. stop isotropic confinment and start axial loading)
Frequently Asked Questions
The initial positioning of spheres is done by generating random (x,y,z) in a box and checking if a sphere of radius R (R also randomly generated with respect to a uniform distribution between mean*(1-std_dev) and mean*(1+std_dev) can be inserted at this location without overlaping with others.
If the sphere overlaps, new (x,y,z)’s are generated until a free position for the new sphere is found. This explains the message you have: after 3000 trial-and-error, the sphere couldn’t be placed, and the algorithm stops.
You get the message above if you try to generate an initialy dense packing, which is not possible with this algorithm. It can only generate clouds. You should keep the default value of porosity (n~0.7), or even increase if it is still to low in some cases. The dense state will be obtained in the second step (compaction, see below).
Both algorithm needs numerical parameters to prevent instabilities. For instance, with the method (1) maxWallVelocity is the maximum wall velocity, with method (2) finalMaxMultiplier is the max value of the multiplier applied on sizes at each iteration (always something like 1.00001).
The control of stress on a boundary is based on the total stiffness K of all contacts between the packing and this boundary. In short, at each step, displacement=stress_error/K. This algorithm is implemented in TriaxialStressController, and the control itself is in TriaxialStressController::ControlExternalStress. The control can be turned off independently for each boundary, using the flags wall_XXX_activated, with XXX∈{top, bottom, left, right, back, front}. The imposed sress is a unique value (sigma_iso) for all directions if TriaxialStressController.isAxisymetric, or 3 independent values sigma1, sigma2, sigma3.
The friction during the compaction (whether you are using the expansion method or the compression one for the specimen generation) can be anything between 0 and the final value used during the Triaxial phase. Note that higher friction than the final one would result in volumetric collapse at the beginning of the test. The purpose of using a different value of friction during this phase is related to the fact that the final porosity you get at the end of the sample generation essentially depends on it as well as on the assumed Particle Size Distribution. Changing the initial value of friction will get to a different value of the final porosity.
This internal variable (updated automatically) is true each N timesteps (with N=radiusControlInterval). For other timesteps, there is no expansion. Cycling without expanding is just a way to speed up the simulation, based on the idea that 1% increase each 10 iterations needs less operations than 0.1% at each iteration, but will give similar results.
The value of unbalanced force (dimensionless) is expected to reach low value (i.e. identifying a static-equilibrium condition for the specimen) only at the end of the compaction phase. The code is not aiming at simulating a quasistatic isotropic compaction process, it is only giving a stable packing at the end of it.
A code that is added to output filenames.
Value of unbalanced force for which the system is considered stable. Used in conditionals to switch between loading stages.
Do we just want to generate a stable packing under isotropic pressure (false) or do we want the triaxial loading to start automatically right after compaction stage (true)?
freeze the simulation when conditions are reached (don’t activate this if you want to be able to run/stop from Qt GUI)
auto adjust the isotropic stress state from TriaxialTest::sigmaIsoCompaction to TriaxialTest::sigmaLateralConfinement if they have different values. See docs for TriaxialCompressionEngine::autoUnload
FIXME : what is that?
Friction angle [°] of boundaries contacts.
Ratio of shear vs. normal contact stiffness for boxes.
Stiffness of boxes.
Friction angle [°] of spheres during compaction (different values result in different porosities)]. This value is overridden by TriaxialTest::sphereFrictionDeg before triaxial testing.
Coefficient of Cundal-Non-Viscous damping (applied on on the 3 components of forces)
Coefficient of Cundal-Non-Viscous damping (applied on on the 3 components of torques)
Max time-step. Used as initial value if defined. Latter adjusted by the time stepper.
density of spheres
Return dictionary of attributes.
Use facets for boundaries (not tested)
max multiplier of diameters during internal compaction (secondary precise adjustment)
string that contains some subset (max. 2) of {‘x’,’y’,’z’} ; contains axes will have box dimension hardcoded, even if box is scaled as mean_radius is prescribed: scaling will be applied on the rest.
Generate scene, save to given file
File with positions and sizes of spheres.
flag for choosing between moving boundaries or increasing particles sizes during the compaction stage.
Generate scene, save to temporary file and load immediately
Lower corner of the box.
max multiplier of diameters during internal compaction (initial fast increase)
max velocity of boundaries. Usually useless, but can help stabilizing the system in some cases.
Do not create any files during run (.xml, .spheres, wall stress records)
Number of generated spheres.
interval between size changes when growing spheres.
Mean radius. If negative (default), autocomputed to as a function of box size and TriaxialTest::numberOfGrains
Normalized standard deviation of generated sizes.
interval between file outputs
Confining stress during isotropic compaction.
Lateral stress during triaxial loading. An isotropic unloading is performed if the value is not equal to TriaxialTest::sigmaIsoCompaction.
Friction angle [°] of spheres assigned just before triaxial testing.
Ratio of shear vs. normal contact stiffness for spheres.
Stiffness of spheres.
Strain rate in triaxial loading.
thickness of boundaries. It is arbitrary and should have no effect
interval for GlobalStiffnessTimeStepper
Update object attributes from given dictionary
Upper corner of the box.
Make boundaries larger than the packing to make sure spheres don’t go out during deformation.
interval for updating the stiffness of sample/boundaries contacts
Use walls for boundaries (not tested)
Class responsible for rendering scene on OpenGL devices.
Color of the background canvas (RGB)
Color of the periodic cell (RGB).
Activate/deactivate respective clipping planes
Position and orientation of clipping planes
Return dictionary of attributes.
Artificially enlarge (scale) dispalcements from bodies’ reference positions by this relative amount, so that they become better visible (independently in 3 dimensions). Disbled if (1,1,1).
Show which degrees of freedom are blocked for each body
Additional rendering components (GlExtraDrawer).
Render objects crossing periodic cell edges by cloning them in multiple places (periodic simulations only).
Show body id’s
Draw wire for all interactions, blue for potential and green for real ones (mostly for debugging)
Render Interaction::geom objects.
Render Interaction::phys objects
If rendering interactions, use only wires to represent them.
Turn light 1 on.
Turn light 2 on.
Per-color intensity of secondary light (RGB).
Position of secondary OpenGL light source in the scene.
Per-color intensity of primary light (RGB).
Position of OpenGL light source in the scene.
Bitmask for showing only bodies where ((mask & Body::mask)!=0)
Render the scene in the current OpenGL context.
Artificially enlarge (scale) rotations of bodies relative to their reference orientation, so the they are better visible.
Id of particle that was selected by the user.
Make current positions and orientation reference for scaleDisplacements and scaleRotations.
Update object attributes from given dictionary
Render all bodies with wire only (faster)
Abstract functor for rendering Shape objects.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Renders Box object
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Renders ChainedCylinder object including a shift for compensating flexion.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Renders Cylinder object
Only show wireframe (controlled by glutSlices and glutStacks.
Fix normals for non-wire rendering
Number of sphere slices.
Number of sphere stacks.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Renders Facet object
In wire mode, render normals of facets and edges; facet’s colors are disregarded in that case.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Renders Cylinder object
Only show wireframe (controlled by glutSlices and glutStacks.
Fix normals for non-wire rendering
Number of cylinder slices.
Number of cylinder stacks.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Renders Polyhedra object
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Renders Sphere object
Change discretization level of spheres. quality>1 for better image quality, at the price of more cpu/gpu usage, 0<quality<1 for faster rendering. If mono-color spheres are displayed (Gl1_Sphere::stripes = False), quality mutiplies Gl1_Sphere::glutSlices and Gl1_Sphere::glutStacks. If striped spheres are displayed (Gl1_Sphere::stripes = True), only integer increments are meaningfull : quality=1 and quality=1.9 will give the same result, quality=2 will give finer result.
Only show wireframe (controlled by glutSlices and glutStacks.
In non-wire rendering, show stripes clearly showing particle rotation.
Compute specular light in local eye coordinate system.
Base number of sphere slices, multiplied by Gl1_Sphere::quality before use); not used with stripes (see glut{Solid,Wire}Sphere reference)
Base number of sphere stacks, multiplied by Gl1_Sphere::quality before use; not used with stripes (see glut{Solid,Wire}Sphere reference)
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Renders Tetra object
TODO
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Renders Wall object
Number of divisions of the wall inside visible scene part.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Abstract functor for rendering State objects.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Abstract functor for rendering Bound objects.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Render Axis-aligned bounding box (Aabb).
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Abstract functor for rendering IGeom objects.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Render L3Geom geometry.
Whether to display labels for local axes (x,y,z)
Scale local axes, their reference length being half of the minimum radius.
Width of axes lines, in pixels; not drawn if non-positive
Width of lines for drawing displacements (and rotations for L6Geom); not drawn if non-positive.
Scale local displacements (u - u0); 1 means the true scale, 0 disables drawing local displacements; negative values are permissible.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Render L6Geom geometry.
Scale local rotations (phi - phi0). The default scale is to draw rotation with length equal to minimum radius.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Abstract functor for rendering IPhys objects.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Render CpmPhys objects of interactions.
Show contact line
Numerically show contact damage parameter
[what is this?]
Show shear strain
Show axes of shear plane
Show contact normal
If positive, set the interaction (wire) color based on normalized by
× colorStrainRatio (
= CpmPhys.epsCrackOnset ). Otherwise, color based on the residual strength.
Numerically show normal strain
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Renders NormPhys objects as cylinders of which diameter and color depends on NormPhys::normForce magnitude.
Value of NormPhys.normalForce corresponding to maxDiameter. This value will be increased (but not decreased ) automatically.
If non-zero, only display contacts with negative (-1) or positive (+1) normal forces; if zero, all contacts will be displayed.
Reference (minimum) particle radius; used only if maxRadius is negative. This value will be decreased (but not increased ) automatically. (auto-updated)
Cylinder radius corresponding to the maximum normal force. If negative, auto-updated refRadius will be used instead.
Number of sphere slices; (see glutCylinder reference)
Number of sphere stacks; (see glutCylinder reference)
Value that divides contacts by their normal force into the ``weak fabric’’ and ``strong fabric’‘. This value is set as side-effect by utils.fabricTensor.
If non-zero, only display contacts belonging to the ``weak’’ (-1) or ``strong’’ (+1) fabric.
If maxWeakFn is set, scale radius of the weak fabric by this amount (usually smaller than 1). If zero, 1 pixel line is displayed. Colors are not affected by this value.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Add new scene to Omega, returns its number
Bodies in the current simulation (container supporting index access by id and iteration)
Periodic cell of the current scene (None if the scene is aperiodic).
Return list of all classes deriving from given class, as registered in the class factory
Revert SEGV and ABRT handlers to system defaults.
Current timestep (Δt) value.
Whether a TimeStepper is used for dynamic Δt control. See dt on how to enable/disable TimeStepper.
Whether a TimeStepper is amongst O.engines, activated or not.
EnergyTracker of the current simulation. (meaningful only with O.trackEnergy)
List of engines in the simulation (Scene::engines).
Disable SEGV handler and exit, optionally with given status number.
Filename under which the current simulation was saved (None if never saved).
Counter for number of syncs in ForceContainer, for profiling purposes.
ForceContainer (forces, torques, displacements) in the current simulation.
Interactions in the current simulation (container supporting index acces by either (id1,id2) or interactionNumber and iteration)
Tells whether the first class derives from the second one (both given as strings).
Get current step number
Return instance of engine/functor with the given label. This function shouldn’t be called by the user directly; every ehange in O.engines will assign respective global python variables according to labels.
For example:
O.engines=[InsertionSortCollider(label=’collider’)]
collider.nBins=5 # collider has become a variable after assignment to O.engines automatically
Load simulation from file. The file should be saved in the same version of Yade, otherwise compatibility is not guaranteed.
Load simulation previously stored in memory by saveTmp. mark optionally distinguishes multiple saved simulations
Return list of all memory-saved simulations.
Shared materials; they can be accessed by id or by label
MiscParams in the simulation (Scene::mistParams), usually used to save serializables that don’t fit anywhere else, like GL functors
Get maximum number of threads openMP can use.
Stop simulation execution. (May be called from within the loop, and it will stop after the current step).
Get/set whether the scene is periodic or not (True/False).
Return list of all plugins registered in the class factory.
Return clock (human world) time the simulation has been running.
Reload current simulation
Reset simulations completely (including another scenes!).
Reset all scenes.
Reset current scene.
Reset current scene.
Reset simulation time: step number, virtual and real time. (Doesn’t touch anything else, including timings).
Run the simulation. nSteps how many steps to run, then stop (if positive); wait will cause not returning to python until simulation will have stopped.
Run given engine exactly once; simulation time, step number etc. will not be incremented (use only if you know what you do).
Whether background thread is currently running a simulation.
Save current simulation to file (should be .xml or .xml.bz2). The file should be loaded in the same version of Yade, otherwise compatibility is not guaranteed.
Save simulation to memory (disappears at shutdown), can be loaded later with loadTmp. mark optionally distinguishes different memory-saved simulations.
Return the entire scene as a string. Equivalent to using O.save(...) except that the scene goes to a string instead of a file. (see also stringToScene())
Return current calculation speed [iter/sec].
Advance the simulation by one step. Returns after the step will have finished.
Get/set number of iteration after which the simulation will stop.
Load simulation from a string passed as argument (see also sceneToString).
Get the current subStep number (only meaningful if O.subStepping==True); -1 when outside the loop, otherwise either 0 (O.subStepping==False) or number of engine to be run (O.subStepping==True)
Get/set whether subStepping is active.
Switch to alternative simulation (while keeping the old one). Calling the function again switches back to the first one. Note that most variables from the first simulation will still refer to the first simulation even after the switch (e.g. b=O.bodies[4]; O.switchScene(); [b still refers to the body in the first simulation here])
Switch to defined scene. Default scene has number 0, other scenes have to be created by addScene method.
Tags (string=string dictionary) of the current simulation (container supporting string-index access/assignment)
Return virtual (model world) time of the simulation.
Globally enable/disable timing services (see documentation of the timing module).
Return unique name of file in temporary directory which will be deleted when yade exits.
When energy tracking is enabled or disabled in this simulation.
Don’t return until the simulation will have been paused. (Returns immediately if not running).
Add body b (or a list of bodies) to an existing clump c. c must be clump and b may not be a clump member of c. Clump masses and inertia are adapted automatically (for details see clump()).
See examples/clumps/addToClump-example.py for an example script.
Note
If b is a clump itself, then all members will be added to c and b will be deleted. If b is a clump member of clump d, then all members from d will be added to c and d will be deleted. If you need to add just clump member b, release this member from d first.
Append one Body instance, return its id.
Append given list of bodies as a clump (rigid aggregate); returns a tuple of (clumpId,[memberId1,memberId2,...]). Clump masses and inertia are adapted automatically (for details see clump()).
Remove all bodies (interactions not checked)
Clump given bodies together (creating a rigid aggregate); returns clumpId. Clump masses and inertia are adapted automatically (default with integrateInertia=True). If clump members are overlapping this is done by integration/summation over mass points using a regular grid of cells (number of grid cells in one direction is defined as , where
is minimum clump member radius). For non-overlapping members inertia of the clump is the sum of inertias from members. If integrateInertia=False sum of inertias from members is used (faster, but inaccurate).
Erase body with the given id; all interaction will be deleted by InteractionLoop in the next step.
Returns roundness coefficient RC = R2/R1. R1 is the theoretical radius of a sphere, with same volume as clump. R2 is the minimum radius of a sphere, that imbeds clump. If just spheres are present RC = 1. If clumps are present 0 < RC < 1. Bodies can be excluded from the calculation by giving a list of ids: O.bodies.getRoundness([ids]).
See examples/clumps/replaceByClumps-example.py for an example script.
Release body b from clump c. b must be a clump member of c. Clump masses and inertia are adapted automatically (for details see clump()).
See examples/clumps/releaseFromClump-example.py for an example script.
Note
If c contains only 2 members b will not be released and a warning will appear. In this case clump c should be erased.
Replace spheres by clumps using a list of clump templates and a list of amounts; returns a list of tuples: [(clumpId1,[memberId1,memberId2,...]),(clumpId2,[memberId1,memberId2,...]),...]. A new clump will have the same volume as the sphere, that was replaced. Clump masses and inertia are adapted automatically (for details see clump()).
O.bodies.replaceByClumps( [utils.clumpTemplate([1,1],[.5,.5])] , [.9] ) #will replace 90 % of all standalone spheres by ‘dyads’
See examples/clumps/replaceByClumps-example.py for an example script.
Access to interactions of simulation, by using
id’s of both Bodies of the interactions, e.g. O.interactions[23,65]
iteraction over the whole container:
for i in O.interactions: print i.id1,i.id2
Note
Iteration silently skips interactions that are not real.
Remove all interactions, and invalidate persistent collider data (if the collider supports it).
Return number of interactions that are “real”, i.e. they have phys and geom.
Erase one interaction, given by id1, id2 (internally, requestErase is called – the interaction might still exist as potential, if the Collider decides so).
Return n-th interaction from the container (usable for picking random interaction).
Return list of real interactions of given body.
Return list of all (real as well as non-real) interactions of given body.
Apply force on body (accumulates).
# If permanent=false (default), the force applies for one iteration, then it is reset by ForceResetter. # If permanent=true, it persists over iterations, until it is overwritten by another call to addF(id,f,True) or removed by reset(resetAll=True). The permanent force on a body can be checked with permF(id).
Apply displacement on body (accumulates).
Apply rotation on body (accumulates).
Apply torque on body (accumulates).
# If permanent=false (default), the torque applies for one iteration, then it is reset by ForceResetter. # If permanent=true, it persists over iterations, until it is overwritten by another call to addT(id,f,True) or removed by reset(resetAll=True). The permanent torque on a body can be checked with permT(id).
Force applied on body. For clumps in openMP, synchronize the force container with sync=True, else the value will be wrong.
Check wether permanent forces are present.
Deprecated alias for t (torque).
Displacement applied on body.
read the value of permanent force on body (set with setPermF()).
read the value of permanent torque on body (set with setPermT()).
Reset the force container, including user defined permanent forces/torques. resetAll=False will keep permanent forces/torques unchanged.
Rotation applied on body.
Number of synchronizations of ForceContainer (cummulative); if significantly higher than number of steps, there might be unnecessary syncs hurting performance.
Torque applied on body. For clumps in openMP, synchronize the force container with sync=True, else the value will be wrong.
Container for Materials. A material can be accessed using
Add new shared Material; changes its id and return it.
Return id of material, given its label.
Object comprising the whole simulation.
Whether the convention is that compression has negative sign (set by IGeomFunctor).
Return dictionary of attributes.
Used, when new body is added to the scene.
Current timestep for integration.
Various flags of the scene; 1 (Scene::LOCAL_COORDS): use local coordinate system rather than global one for per-interaction quantities (set automatically from the functor).
Whether periodic boundary conditions are active.
Current iteration (computational step) number
Whether local coordianate system is used on interactions (set by IGeomFunctor).
Id of body that is selected by the user
Current calculation speed [iter/s]
Iteration after which to stop the simulation.
Time after which to stop the simulation
Number of sub-step; not to be changed directly. -1 means to run loop prologue (cell integration), 0…n-1 runs respective engines (n is number of engines), n runs epilogue (increment step number and time.
Whether we currently advance by one engine in every step (rather than by single run through all engines).
Arbitrary key=value associations (tags like mp3 tags: author, date, version, description etc.)
Simulation time (virtual time) [s]
Whether energies are being traced.
Update object attributes from given dictionary
Parameters of periodic boundary conditions. Only applies if O.isPeriodic==True.
Return dictionary of attributes.
Returns deformation gradient tensor of the cell deformation (http://en.wikipedia.org/wiki/Finite_strain_theory)
Returns Eulerian-Almansi strain tensor of the cell (http://en.wikipedia.org/wiki/Finite_strain_theory)
Returns left Cauchy-Green deformation tensor of the cell (http://en.wikipedia.org/wiki/Finite_strain_theory)
Returns Lagrangian strain tensor of the cell (http://en.wikipedia.org/wiki/Finite_strain_theory)
Returns left (spatial) stretch tensor of the cell (matrix from polar decomposition
)
Returns orthogonal matrix and symmetric positive semi-definite matrix
as polar decomposition of deformation gradient
of the cell (
)
Returns right Cauchy-Green deformation tensor of the cell (http://en.wikipedia.org/wiki/Finite_strain_theory)
Returns right (material) stretch tensor of the cell (matrix from polar decomposition
)
Returns rotation of the cell (orthogonal matrix from polar decomposition
)
Returns small strain tensor of the cell (http://en.wikipedia.org/wiki/Finite_strain_theory)
Base cell vectors (columns of the matrix), updated at every step from velGrad (trsf accumulates applied velGrad transformations). Setting hSize during a simulation is not supported by most contact laws, it is only meant to be used at iteration 0 before any interactions have been created.
Deform (velGrad) the cell homothetically, by adjusting positions and velocities of bodies. The velocity change is obtained by deriving the expression v=∇v.x, where ∇v is the macroscopic velocity gradient, giving in an incremental form: Δv=Δ ∇v x + ∇v Δx. As a result, velocities are modified as soon as velGrad changes, according to the first term: Δv(t)=Δ ∇v x(t), while the 2nd term reflects a convective term: Δv’= ∇v v(t-dt/2).
see Cell.velGrad.
hSize from the previous step, used in the definition of relative velocity across periods.
Velocity gradient in the previous step.
Reference cell configuration, only used with OpenGLRenderer.dispScale. Updated automatically when hSize or trsf is assigned directly; also modified by utils.setRefSe3 (called e.g. by the Reference button in the UI).
Reference size of the cell (lengths of initial cell vectors, i.e. column norms of hSize).
Note
Modifying this value is deprecated, use setBox instead.
Apply shear (cell skew+rot) on the point
Current skew+rot transformation (no resize)
Current size of the cell, i.e. lengths of the 3 cell lateral vectors contained in Cell.hSize columns. Updated automatically at every step.
Current transformation matrix of the cell, obtained from time integration of Cell.velGrad.
Apply inverse shear on the point (removes skew+rot of the cell)
Inverse of the current skew+rot transformation (no resize)
Update object attributes from given dictionary
Velocity gradient of the transformation; used in NewtonIntegrator. Values of velGrad accumulate in trsf at every step.
NOTE: changing velGrad at the beginning of a simulation loop would lead to inacurate integration for one step, as it should normaly be changed after the contact laws (but before Newton). To avoid this problem, assignment is deferred automatically. The target value typed in terminal is actually stored in Cell.nextVelGrad and will be applied right in time by Newton integrator.
true when velGrad has been changed manually (see also Cell.nextVelGrad)
Current volume of the cell.
Transform an arbitrary point into a point in the reference cell
Wrap point inside the reference cell, assuming the cell has no skew+rot.
Basic execution unit of simulation, called from the simulation loop (O.engines)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Parameters of periodic boundary conditions. Only applies if O.isPeriodic==True.
Return dictionary of attributes.
Returns deformation gradient tensor of the cell deformation (http://en.wikipedia.org/wiki/Finite_strain_theory)
Returns Eulerian-Almansi strain tensor of the cell (http://en.wikipedia.org/wiki/Finite_strain_theory)
Returns left Cauchy-Green deformation tensor of the cell (http://en.wikipedia.org/wiki/Finite_strain_theory)
Returns Lagrangian strain tensor of the cell (http://en.wikipedia.org/wiki/Finite_strain_theory)
Returns left (spatial) stretch tensor of the cell (matrix from polar decomposition
)
Returns orthogonal matrix and symmetric positive semi-definite matrix
as polar decomposition of deformation gradient
of the cell (
)
Returns right Cauchy-Green deformation tensor of the cell (http://en.wikipedia.org/wiki/Finite_strain_theory)
Returns right (material) stretch tensor of the cell (matrix from polar decomposition
)
Returns rotation of the cell (orthogonal matrix from polar decomposition
)
Returns small strain tensor of the cell (http://en.wikipedia.org/wiki/Finite_strain_theory)
Base cell vectors (columns of the matrix), updated at every step from velGrad (trsf accumulates applied velGrad transformations). Setting hSize during a simulation is not supported by most contact laws, it is only meant to be used at iteration 0 before any interactions have been created.
Deform (velGrad) the cell homothetically, by adjusting positions and velocities of bodies. The velocity change is obtained by deriving the expression v=∇v.x, where ∇v is the macroscopic velocity gradient, giving in an incremental form: Δv=Δ ∇v x + ∇v Δx. As a result, velocities are modified as soon as velGrad changes, according to the first term: Δv(t)=Δ ∇v x(t), while the 2nd term reflects a convective term: Δv’= ∇v v(t-dt/2).
see Cell.velGrad.
hSize from the previous step, used in the definition of relative velocity across periods.
Velocity gradient in the previous step.
Reference cell configuration, only used with OpenGLRenderer.dispScale. Updated automatically when hSize or trsf is assigned directly; also modified by utils.setRefSe3 (called e.g. by the Reference button in the UI).
Reference size of the cell (lengths of initial cell vectors, i.e. column norms of hSize).
Note
Modifying this value is deprecated, use setBox instead.
Apply shear (cell skew+rot) on the point
Current skew+rot transformation (no resize)
Current size of the cell, i.e. lengths of the 3 cell lateral vectors contained in Cell.hSize columns. Updated automatically at every step.
Current transformation matrix of the cell, obtained from time integration of Cell.velGrad.
Apply inverse shear on the point (removes skew+rot of the cell)
Inverse of the current skew+rot transformation (no resize)
Update object attributes from given dictionary
Velocity gradient of the transformation; used in NewtonIntegrator. Values of velGrad accumulate in trsf at every step.
NOTE: changing velGrad at the beginning of a simulation loop would lead to inacurate integration for one step, as it should normaly be changed after the contact laws (but before Newton). To avoid this problem, assignment is deferred automatically. The target value typed in terminal is actually stored in Cell.nextVelGrad and will be applied right in time by Newton integrator.
true when velGrad has been changed manually (see also Cell.nextVelGrad)
Current volume of the cell.
Transform an arbitrary point into a point in the reference cell
Wrap point inside the reference cell, assuming the cell has no skew+rot.
Get timing data as list of tuples (label, execTime[nsec], execCount) (one tuple per checkpoint)
Reset timing information
Performing arbitrary OpenGL drawing commands; called from OpenGLRenderer (see OpenGLRenderer.extraDrawers) once regular rendering routines will have finished.
This class itself does not render anything, derived classes should override the render method.
Deactivate the object (on error/exception).
Return dictionary of attributes.
Update object attributes from given dictionary
Dispatcher calling functors based on received argument type(s).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Return functor that would be dispatched for given argument(s); None if no dispatch; ambiguous dispatch throws.
Return dictionary with contents of the dispatch matrix.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Functors associated with this dispatcher.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Engine for running other Engine in parallel.
object __init__(tuple args, dict kwds)
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
List of lists of Engines; each top-level group will be run in parallel with other groups, while Engines inside each group will be run sequentially, in given order.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Dispatcher calling functors based on received argument type(s).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Return functor that would be dispatched for given argument(s); None if no dispatch; ambiguous dispatch throws.
Return dictionary with contents of the dispatch matrix.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Functors associated with this dispatcher.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Function-like object that is called by Dispatcher, if types of arguments match those the Functor declares to accept.
Ordered list of types (as strings) this functor accepts.
Return dictionary of attributes.
Textual label for this object; must be a valid python identifier, you can refer to it directly from python.
Detailed information about timing inside the Dispatcher itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Return dictionary of attributes.
Update object attributes from given dictionary
Find an instance of LawTester and show visually its data.
Deactivate the object (on error/exception).
Return dictionary of attributes.
Update object attributes from given dictionary
Dispatcher calling functors based on received argument type(s).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Return functor that would be dispatched for given argument(s); None if no dispatch; ambiguous dispatch throws.
Return dictionary with contents of the dispatch matrix.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Functors associated with this dispatcher.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Class matching pair of ids to return pre-defined (for a pair of ids defined in matches) or derived value (computed using algo) of a scalar parameter. It can be called (id1, id2, val1=NaN, val2=NaN) in both python and c++.
Note
There is a converter from python number defined for this class, which creates a new MatchMaker returning the value of that number; instead of giving the object instance therefore, you can only pass the number value and it will be converted automatically.
Alogorithm used to compute value when no match for ids is found. Possible values are
The following algo algorithms do not require meaningful input values in order to work:
Compute algo value for val1 and val2, using algorithm specified by algo.
Return dictionary of attributes.
Array of (id1,id2,value) items; queries matching id1 + id2 or id2 + id1 will return value
Update object attributes from given dictionary
Dispatcher calling functors based on received argument type(s).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Return functor that would be dispatched for given argument(s); None if no dispatch; ambiguous dispatch throws.
Return dictionary with contents of the dispatch matrix.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Functors associated with this dispatcher.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Dispatcher calling functors based on received argument type(s).
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Return functor that would be dispatched for given argument(s); None if no dispatch; ambiguous dispatch throws.
Return dictionary with contents of the dispatch matrix.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Functors associated with this dispatcher.
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Render boxed read from file
File to read boxes from; ascii files with x0 y0 z0 x1 y1 z1 c records, where c is an integer specifying fill (0 for wire, 1 for filled).
Deactivate the object (on error/exception).
Return dictionary of attributes.
Range of fill indices that will be rendered.
Range of fill indices that will be filled.
Range of levels that will be rendered.
Do not fill 0-fill boxed (those that are further subdivided)
Update object attributes from given dictionary
Engine dispatching control to its associated functors, based on types of argument it receives. This abstract base class provides no functionality in itself.
If true, this engine will not run at all; can be used for making an engine temporarily deactivated and only resurrect it at a later point.
Return dictionary of attributes.
Cummulative count this engine was run (only used if O.timingEnabled==True).
Cummulative time this Engine took to run (only used if O.timingEnabled==True).
Textual label for this object; must be valid python identifier, you can refer to it directly from python.
Number of threads to be used in the engine. If ompThreads<0 (default), the number will be typically OMP_NUM_THREADS or the number N defined by ‘yade -jN’ (this behavior can depend on the engine though). This attribute will only affect engines whose code includes openMP parallel regions (e.g. InteractionLoop). This attribute is mostly useful for experiments or when combining ParallelEngine with engines that run parallel regions, resulting in nested OMP loops with different number of threads at each level.
Detailed information about timing inside the Engine itself. Empty unless enabled in the source code and O.timingEnabled==True.
Update object attributes from given dictionary
Storage for tracing energies. Only to be used if O.trackEnergy is True.
Clear all stored values.
Return dictionary of attributes.
Energy values, in linear array
Return contents as list of (name,value) tuples.
Return defined energies.
Return sum of all energies.
Update object attributes from given dictionary