Field outputs

Relevant module: edelweissfe.utils.fieldoutput

FieldOutputs store all kind of analysis results, and are defined via the keyword *fieldOutput. All fieldOutputs are accessable to all outputmanagers at the end of each increment, step and job. Furthermore, they can be exported to *.csv files at the end of the analysis job.

Attention

If the results are exported to a .csv file with enabled “saveHistory”, the time History is automatically appended to the .csv file”

Options:

Option	Type / Default	Description
perNode	tuple, default=()	Create node-based field output.
perElement	tuple, default=()	Create element-based field output.
fromExpression	tuple, default=()	Create field output from expression.

Keyword: perNode

Option	Type / Default	Description
name	str (required)	Name of the field output.
field	str (required)	Field of the result.
result	str (required)	Result name.
elSet	str, default=None	Element set.
nSet	str, default=None	Node set.
rigidBody	str, default=None	Rigid body (as registered in model.rigidBodies).
saveHistory	bool, default=False	Save complete History or only last (increment) result
f(x)	str, default=None	Function to apply in each increment.
f_export(x)	str, default=None	Function to apply on final result (table).
export	str, default=None	Export the field output to a file at the end of the job.

Keyword: perElement

Option	Type / Default	Description
name	str (required)	Name of the field output.
elSet	str (required)	Element set.
result	str (required)	Result name.
quadraturePoint	str (required)	Integer or slice.
saveHistory	bool, default=False	Save complete History or only last (increment) result
f(x)	str, default=None	Function to apply in each increment.
f_export(x)	str, default=None	Function to apply on final result (table).
export	str, default=None	Export the field output to a file at the end of the job.

Keyword: fromExpression

Option	Type / Default	Description
name	str (required)	Name of the field output.
elSet	str, default=None	Element set.
nSet	str, default=None	Node set.
expression	str (required)	Expression for retrieving field output.
saveHistory	bool, default=False	Save complete History or only last (increment) result
f(x)	str, default=None	Function to apply in each increment.
f_export(x)	str, default=None	Function to apply on final result (table).
export	str, default=None	Export the field output to a file at the end of the job.

Example file: ../../../testfiles/marmot/FieldOutput/fieldOutput.inp

*material, name=modleon, id=modleon
30000.0, 0.15 ,13 , 47.4 ,  55   , 4.74 , 0.85 , 0.12 , 0.003, 2.0, 0.000001, 15.0, 0.10, 1

*section, name=section1, material=modleon, type=plane
all

*job, name=cps4job, domain=2d, solver=NIST
*solver, solver=NIST, name=theSolver

*modelGenerator, generator=planeRectQuad, name=gen
x0=0, l=50
y0=0, h=100
elType=CPS4
nX=20
nY=20

*fieldOutput,
>>perNode, name=displacement,              nSet=all,           field=displacement, result=U
>>perNode, name=displacement_top,          nSet=gen_top,       field=displacement, result=U, saveHistory=True
>>perNode, name=displacement_top_mean_1,   nSet=gen_top,       field=displacement, result=U, saveHistory=True, f(x)='mean( x[:,1])'
>>perNode, name=RF_bottom,                 nSet=gen_bottom,    field=displacement, result=P, saveHistory=True
>>perNode, name=RF_bottom_sum_1,           nSet=gen_bottom,    field=displacement, result=P, saveHistory=True, f(x)='sum(x[:,1])'
>>perElement, name=stress,                    elSet=all,          result=stress,      quadraturePoint=1
>>perElement, name=stress_top,                elSet=gen_top,      result=stress,      quadraturePoint=1
>>perElement, name=alphaP,                    elSet=all,          result=alphaP,      quadraturePoint=1

** *output, type=meshPlot,
** figure=1, axSpec=221, create=perNode,       fieldOutput=displacement,       f(x)='x[:,1]', label=U2
** figure=1, axSpec=222, create=perElement,    fieldOutput=stress,             f(x)='x[:,1]', label=S2
** figure=1, axSpec=212, c=red, ls=dashed, create=xyData, x=time, y=RF_bottom, f(y)='sum ( y[:,:,1], axis=1)', label=Sum(RF)

*output, type=ensight, name=myEnsightExport
>>perNode, fieldOutput=displacement
>>perElement, fieldOutput=stress
>>perElement, fieldOutput=alphaP
>>configuration, overwrite=yes


*output, type=monitor, name=myNodeMonitor
fieldOutput=RF_bottom, f(x)='sum ( x[:,1] )'

*output, type=plotAlongPath, name=myNodeSetPlotter
figure=2, nSet=gen_top, fieldOutput=RF_bottom, f(x)='x[:,1]', label=P along bottom, normalize=True

*step, solver=theSolver
maxInc=1e-1, minInc=1e-8, maxNumInc=1000, maxIter=25, stepLength=100
>>options, name=myEnsightExport, intermediateSaveInterval=2
>>options, name=theSolver, extrapolation=linear
>>dirichlet, name=1,          nSet=gen_leftBottom,    field=displacement, 2=0.0, 1=0.0
>>dirichlet, name=bottom,     nSet=gen_bottom,        field=displacement, 2=0,
>>nodeforces, name=top,       nSet=gen_top,           field=displacement, 2=-80, f(t)=t**2

Module documentation:

class edelweissfe.utils.fieldoutput.FieldOutputController(model, journal)

The central module for managing field outputs, which can be used by output managers.

Methods

addExpressionFieldOutput(associatedSet, ...)	Add a new FieldOutput entry to be computed during the simulation
addPerElementFieldOutput(name, elSet[, ...])	Add a new FieldOutput entry to be computed during the simulation
addPerNodeFieldOutput(name, nodeField[, ...])	Add a new FieldOutput entry to be computed during the simulation
addRigidBodyFieldOutput(name, rigidBody, field)	Add a new FieldOutput entry operating directly on a RigidBody.
finalizeIncrement()	Finalize all field outputs at the end of an increment.
finalizeJob()	Finalize all field outputs at the end of a job.
finalizeStep()	Finalize all field outputs at the end of a step.
initializeStep(step)	Initalize an step.

initializeJob

Parameters:

• model (FEModel)

• journal (Journal)

addExpressionFieldOutput(associatedSet, theExpression, name, saveHistory=False, f_x=None, export=None, fExport_x=None)

Add a new FieldOutput entry to be computed during the simulation

Parameters:

• associatedSet (set) – The associated set of nodes or elements.

• theExpression (Callable) – The expression to be evaluated.

• name (str) – The name of this FieldOutput.

• saveHistory – Save the complete history or only the last result.

• f_x (Callable) – Apply a math function on the results.

• export (str) – Export the results to a file.

• fExport_x (Callable) – Apply a math function on the results before exporting.

addPerElementFieldOutput(name, elSet, result=None, saveHistory=False, f_x=None, export=None, fExport_x=None, quadraturePoints=int | slice | list[int])

Add a new FieldOutput entry to be computed during the simulation

Parameters:

• name (str) – The name of this FieldOutput.

• elSet (ElementSet) – The ElementSet on which this FieldOutput should operate.

• result (str) – The name of the result, which is provided by the Elements in the ElementSet.

• journal – The Journal instance for logging purposes.

• saveHistory – Save the complete history or only the last result.

• f_x (Callable) – Apply a math function on the results.

• export (str) – Export the results to a file.

• fExport_x (Callable) – Apply a math function on the results before exporting.

• quadraturePoints – The indices of quadrature points for which the results should be extracted.

addPerNodeFieldOutput(name, nodeField, result=None, saveHistory=False, f_x=None, export=None, fExport_x=None)

Add a new FieldOutput entry to be computed during the simulation

Parameters:

• name (str) – The name of this FieldOutput.

• nodeField (NodeField) – The NodeField on which this FieldOutput should operate.

• result (str) – The name of the result entry in the NodeField.

• journal – The Journal instance for logging purposes.

• saveHistory – Save the complete history or only the last result.

• f_x (Callable) – Apply a math function on the results.

• export (str) – Export the results to a file.

• fExport_x (Callable) – Apply a math function on the results before exporting.

addRigidBodyFieldOutput(name, rigidBody, field, saveHistory=False, f_x=None, export=None, fExport_x=None)

Add a new FieldOutput entry operating directly on a RigidBody.

Parameters:

• name (str) – The name of this FieldOutput.

• rigidBody – The RigidBody on which this FieldOutput should operate.

• field (str) – The name of the field to retrieve from the rigid body (e.g., “displacement”).

• saveHistory – Save the complete history or only the last result.

• f_x (Callable) – Apply a math function on the results.

• export (str) – Export the results to a file.

• fExport_x (Callable) – Apply a math function on the results before exporting.

finalizeIncrement()

Finalize all field outputs at the end of an increment.

finalizeJob()

Finalize all field outputs at the end of a job.

finalizeStep()

Finalize all field outputs at the end of a step.

initializeStep(step)

Initalize an step.

Parameters:

step – The step information.

class edelweissfe.utils.fieldoutput._FieldOutputBase(name, model, journal, saveHistory=False, f_x=None, export=None, fExport_x=None, reshape_to_dimensions=None)

Entity of a fieldOutput request. Carries the history or the latest result.

Parameters:

• name (str) – The name of this FieldOutput.

• model (FEModel) – A dictionary containing the model tree.

• journal (Journal) – The journal object for logging.

• saveHistory (bool) – Save the complete history or only the last result.

• f_x (Callable) – Apply a math function on the results.

• export (str) – Export the results to a file.

• fExport_x (Callable) – Apply a math function on the results before exporting.

• reshape_to_dimensions (int) – Reshape the result to a specific number of dimensions.

Attributes:

resultVersion

The number of changes of result so far – its updates and a restart’s load: the key of whatever is derived from the result and kept until it changes (e.g.

Methods

finalizeIncrement()	Finalize an increment, i.e. store the current results.
finalizeStep()	Bring the stored result up to date with the model's final state, if the step advanced past the increment that stored the current one.
getLastResult()	Get the last result, no matter if the history is stored or not.
getRestartData()	The history this field output carries from one increment to the next -- the times, the results (flattened, since refinement changes their shape), and how much of the export file belongs to the run.
getResultHistory()	Get the history.
getTimeHistory()	Get the time history.
hasResult()	Whether a result was computed yet: False before the first converged increment.
initializeJob()	Initalize everything.
initializeStep(step)	Write the current (just-updated) result as the first row of this step.
setRestartData(data)	Restore the state getRestartData() returned.
setResults(values)	Modifies a result at it's origin, if possible.
writeLastResult()	Update file output.

finalizeJob

finalizeIncrement()

Finalize an increment, i.e. store the current results.

finalizeStep()

Bring the stored result up to date with the model’s final state, if the step advanced past the increment that stored the current one.

The output managers write one final frame at the end of a step, from the result stored here – see the Ensight export’s finalizeStep, which writes whenever time has advanced since its last frame. If the topology changed after that result was stored, which is exactly what live h-adaptivity does between two output increments, the result belongs to a different mesh than the one being written and the export rejects it:

Variable displacement result size (32) does not match the number of nodes (141)

Guarded on time having advanced, by the same rule the managers use, so the two decisions cannot disagree: whenever a manager writes a final frame the result behind it has just been refreshed, and a step ending on an increment that already stored a result does not store a second one at the same time and duplicate the last point of a history export.

getLastResult()

Get the last result, no matter if the history is stored or not.

Returns:

The result array.

Return type:

np.ndarray

getRestartData()

The history this field output carries from one increment to the next – the times, the results (flattened, since refinement changes their shape), and how much of the export file belongs to the run.

Returns:

The state.

Return type:

dict[str, numpy.ndarray]

getResultHistory()

Get the history. Throws an exception if the history is not stored, or if the stored entries do not all share the same shape (e.g. because the mesh changed under adaptive refinement, in which case a rectangular per-node history is not well-defined).

Returns:

The result history.

Return type:

np.ndarray

getTimeHistory()

Get the time history.

Returns:

The time history.

Return type:

np.ndarray

hasResult()

Whether a result was computed yet: False before the first converged increment.

Returns:

True once getLastResult() has something to return.

Return type:

bool

initializeJob()

Initalize everything. Will also update the results based on the proved start time and solution.

initializeStep(step)

Write the current (just-updated) result as the first row of this step.

Only at the start of a step: a step resumed from a checkpoint already exported this state, as the last row before the checkpoint was written.

setRestartData(data)

Restore the state getRestartData() returned.

Parameters:

data (dict[str, ndarray]) – The state.

setResults(values)

Modifies a result at it’s origin, if possible. Throws an exception if not possible.

Parameters:

values (ndarray) – The values.

writeLastResult()

Update file output.

Parameters:

model – The model tree.

property resultVersion: int

the key of whatever is derived from the result and kept until it changes (e.g. a marker’s selectionStamp()).

A cache key, not state: a resumed run counts from its restart, not from the start of the analysis. It is backed by the change counter _version, which, like the _version of a set, a surface or a node field, a restart need not reproduce.

Type:

The number of changes of result so far – its updates and a restart’s load

class edelweissfe.utils.fieldoutput.ElementFieldOutput(name, elSet, resultName, model, journal, saveHistory=False, f_x=None, export=None, fExport_x=None, quadraturePoints=0)

This is a Element based FieldOutput. It operates on ElementSets.

Parameters:

• name (str) – The name of this FieldOutput.

• elSet (ElementSet) – The ElementSet on which this FieldOutput operates.

• resultName (str) – The name of the result entry in the ElementBase.

• model (FEModel) – The model tree instance.

• journal (Journal) – The journal object for logging.

• saveHistory (bool) – Save the complete history or only the last result.

• f_x (Callable) – Apply a math function on the results.

• export (str) – Export the results to a file.

• fExport_x (Callable) – Apply a math function on the results before exporting.

• quadraturePoints (int | slice | list[int]) – The list of quadrature points for which the results should be extracted.

Attributes:

resultVersion

The number of changes of result so far – its updates and a restart’s load: the key of whatever is derived from the result and kept until it changes (e.g.

Methods

finalizeIncrement()	Finalize an increment, i.e. store the current results.
finalizeStep()	Bring the stored result up to date with the model's final state, if the step advanced past the increment that stored the current one.
getLastResult()	Get the last result, no matter if the history is stored or not.
getRestartData()	The history this field output carries from one increment to the next -- the times, the results (flattened, since refinement changes their shape), and how much of the export file belongs to the run.
getResultHistory()	Get the history.
getTimeHistory()	Get the time history.
hasResult()	Whether a result was computed yet: False before the first converged increment.
initializeJob()	Initalize everything.
initializeStep(step)	Write the current (just-updated) result as the first row of this step.
setRestartData(data)	Restore the state getRestartData() returned.
setResults(values)	Modifies a result at it's origin, if possible.
updateResults(model)	Update the field output.
writeLastResult()	Update file output.

finalizeJob

setResults(values)

Modifies a result at it’s origin, if possible. Throws an exception if not possible.

Parameters:

values (ndarray) – The values.

updateResults(model)

Update the field output. Will use the current solution and reaction vector if result is a nodal result.

Parameters:

model (FEModel) – The model tree.

class edelweissfe.utils.fieldoutput.NodeFieldOutput(name, nodeField, result, model, journal, saveHistory=False, f_x=None, export=None, fExport_x=None)

This is a Node based FieldOutput. It operates on NodeFields.

Parameters:

• name (str) – The name of this FieldOutput.

• nodeField – The NodeField, on which this FieldOutput operates.

• result (str) – The name of the result entry in the NodeField, one of ENTRIES.

• model (FEModel) – The model tree instance.

• journal (Journal) – The journal object for logging.

• saveHistory (bool) – Save the complete history or only the last result.

• f_x (Callable) – Apply a math function on the results.

• export (str) – Export the results to a file.

• fExport_x (Callable) – Apply a math function on the results before exporting.

Raises:

KeyError – If the result is not an entry a node field holds.

Attributes:

resultVersion

The number of changes of result so far – its updates and a restart’s load: the key of whatever is derived from the result and kept until it changes (e.g.

Methods

finalizeIncrement()	Finalize an increment, i.e. store the current results.
finalizeStep()	Bring the stored result up to date with the model's final state, if the step advanced past the increment that stored the current one.
getLastResult()	Get the last result, no matter if the history is stored or not.
getRestartData()	The history this field output carries from one increment to the next -- the times, the results (flattened, since refinement changes their shape), and how much of the export file belongs to the run.
getResultHistory()	Get the history.
getTimeHistory()	Get the time history.
hasResult()	Whether a result was computed yet: False before the first converged increment.
initializeJob()	Initalize everything.
initializeStep(step)	Write the current (just-updated) result as the first row of this step.
setRestartData(data)	Restore the state getRestartData() returned.
setResults(values)	Modifies a result at it's origin, if possible.
updateResults(model)	Update the field output.
writeLastResult()	Update file output.

finalizeJob

updateResults(model)

Update the field output. Will use the current solution and reaction vector if result is a nodal result.

Parameters:

model (FEModel) – The model tree.

class edelweissfe.utils.fieldoutput.RigidBodyFieldOutput(name, rigidBody, field, model, journal, saveHistory=False, f_x=None, export=None, fExport_x=None)

This is a FieldOutput operating directly on a RigidBody.

A rigid body’s visualization (e.g., surface) nodes are not independent degrees of freedom – they are fully determined by the rigid body’s reference point, which is itself part of an ordinary NodeField. This FieldOutput therefore does not operate on a NodeField at all: it queries getVisualizationField(), which computes the requested field directly from the rigid body’s (already solved) kinematics.

Parameters:

• name (str) – The name of this FieldOutput.

• rigidBody – The RigidBody on which this FieldOutput operates.

• field (str) – The name of the field to retrieve from the rigid body (e.g., “displacement”).

• model (FEModel) – The model tree instance.

• journal (Journal) – The journal object for logging.

• saveHistory (bool) – Save the complete history or only the last result.

• f_x (Callable) – Apply a math function on the results.

• export (str) – Export the results to a file.

• fExport_x (Callable) – Apply a math function on the results before exporting.

Attributes:

resultVersion

The number of changes of result so far – its updates and a restart’s load: the key of whatever is derived from the result and kept until it changes (e.g.

Methods

finalizeIncrement()	Finalize an increment, i.e. store the current results.
finalizeStep()	Bring the stored result up to date with the model's final state, if the step advanced past the increment that stored the current one.
getLastResult()	Get the last result, no matter if the history is stored or not.
getRestartData()	The history this field output carries from one increment to the next -- the times, the results (flattened, since refinement changes their shape), and how much of the export file belongs to the run.
getResultHistory()	Get the history.
getTimeHistory()	Get the time history.
hasResult()	Whether a result was computed yet: False before the first converged increment.
initializeJob()	Initalize everything.
initializeStep(step)	Write the current (just-updated) result as the first row of this step.
setRestartData(data)	Restore the state getRestartData() returned.
setResults(values)	Modifies a result at it's origin, if possible.
updateResults(model)	Update the field output from the rigid body's current kinematics.
writeLastResult()	Update file output.

finalizeJob

updateResults(model)

Update the field output from the rigid body’s current kinematics.

Parameters:

model (FEModel) – The model tree.

class edelweissfe.utils.fieldoutput.ExpressionFieldOutput(associatedSet, theExpression, name, model, journal, saveHistory=False, f_x=None, export=None, fExport_x=None)

This is a Node based FieldOutput. It operates on NodeFields.

Parameters:

• associatedSet (OrderedSet | None) – The associated set of nodes or elements.

• theExpression – The expression to be evaluated.

• name (str) – The name of this FieldOutput.

• nodeField – The NodeField, on which this FieldOutput operates.

• result – The name of the result entry in the NodeField.

• model (FEModel) – The model tree instance.

• journal (Journal) – The journal object for logging.

• saveHistory (bool) – Save the complete history or only the last result.

• f_x (Callable) – Apply a math function on the results.

• export (str) – Export the results to a file.

• fExport_x (Callable) – Apply a math function on the results before exporting.

Attributes:

resultVersion

The number of changes of result so far – its updates and a restart’s load: the key of whatever is derived from the result and kept until it changes (e.g.

Methods

finalizeIncrement()	Finalize an increment, i.e. store the current results.
finalizeStep()	Bring the stored result up to date with the model's final state, if the step advanced past the increment that stored the current one.
getLastResult()	Get the last result, no matter if the history is stored or not.
getRestartData()	The history this field output carries from one increment to the next -- the times, the results (flattened, since refinement changes their shape), and how much of the export file belongs to the run.
getResultHistory()	Get the history.
getTimeHistory()	Get the time history.
hasResult()	Whether a result was computed yet: False before the first converged increment.
initializeJob()	Initalize everything.
initializeStep(step)	Write the current (just-updated) result as the first row of this step.
setRestartData(data)	Restore the state getRestartData() returned.
setResults(values)	Modifies a result at it's origin, if possible.
updateResults(model)	Update the field output.
writeLastResult()	Update file output.

finalizeJob

updateResults(model)

Update the field output. Will use the current solution and reaction vector if result is a nodal result.

Parameters:

model (FEModel) – The model tree.