You can import external fields from an output database (.sim) file, also
referred to as a SIM database (see The Output Database). Once imported, the external field can be used
to define distributions, initial conditions, and history-dependent fields (such as
loads, boundary conditions, and predefined fields). External fields can also be used
for node-based submodeling as discussed in About Submodeling. Importing external fields to define
history-dependent fields or node-based submodeling is supported only in Abaqus/Standard.
The external field can be evaluated at nodes, integration points, or element centers.
Conservative mapping is performed automatically if the fields are evaluated at
different locations or if the meshes are different. Importing an external field is
supported for most two-dimensional and three-dimensional continuum elements. When
used to define distributions and initial conditions, external fields are also
supported for three-dimensional conventional shell elements made of a single
material. All field quantities are exchanged in the global Cartesian coordinate
system.
You can import external fields from an output database (.sim) file. You
specify the file name (including the file extension) from which to import external
fields.
Importing an External Field to Define Distributions
You can import external fields to define distributions (see Defining Distributions by Importing Field Data from an Output Database File). A common case is an injection molding manufacturing process where a molten
fibrous material is injected into a mold. You can use injection molding software to
model the manufacturing process and then import the section and material properties
(such as local element orientation and fiber dispersion) into a subsequent
stress-displacement simulation.
Importing an External Field to Adjust Nodal Coordinates
You can import an external field to adjust nodal coordinates.
One common case is to introduce a geometric imperfection into a model for a postbuckling
analysis. In this case, the previous analysis must contain the mode shapes from an
eigenfrequency extraction analysis or nodal displacements from a static or dynamic
analysis. Mapping is performed if the postbuckling analysis uses a dissimilar mesh.
The imperfection is introduced by adjusting the nodal coordinates using the imported
modal shapes or nodal displacements.
Another common case is to adjust nodal coordinates using the deformed configuration from a
previous analysis. In this case, the previous analysis must write the nodal
displacements to a SIM database.
Importing an External Field to Define Initial Conditions
You can import an external field to define initial conditions (see Initial Conditions). A common
case is a geomechanics analysis where you import an initial pore pressure field. You
can compute a pore pressure field using a reservoir code and then import the field
into a subsequent geostatic stress state procedure (see Geostatic Stress State). This procedure is used to compute the
equilibrium state before performing a geotechnical analysis using the coupled pore
fluid diffusion and stress analysis procedure in Abaqus (see Coupled Pore Fluid Diffusion and Stress Analysis). In this case, the reservoir
code must write the pore pressure to a SIM database.
Importing an External Field to Define History-Dependent Fields
You can import external fields to define loads, boundary conditions, and predefined fields. A
common case is a sequential thermal-stress analysis, where you first perform a heat
transfer analysis (see Uncoupled Heat Transfer Analysis) to compute a
temperature field and then import the temperatures into a subsequent
stress-displacement analysis (see Static Stress Analysis).
You can import external fields for the following time-based analysis procedures:
External fields can be imported only in one analysis step per analysis. The analysis step must
be a general analysis (nonperturbation) step. For any subsequent analysis steps, the
applied external fields remain constant based on the last imported values.
Saving Fields in a Previous Analysis
If the previous analysis is an Abaqus analysis, you request the source fields to write to the output database
(.sim) file. When importing external fields over a time
range, you must request output at a frequency to achieve sufficient solution
accuracy (see Controlling the Frequency of Output to the Output Database).
Identifying the Target and Source Fields
When importing an external field, you define the target and source fields. The target field is
the field defined in the current analysis. The source field is the field computed in
a previous analysis. A field is a physical quantity that spans a region. The field
is a node-based field if evaluated at nodes or an element-based field if evaluated
at element centroids, integration points, or surface facets. You define a region by
specifying a node set, an element set, or a surface regardless of where the field is
evaluated.
When a node-based field is imported and an element set from the previous model is
specified, the field is extracted from nodes of the elements in the specified
element set. Similarly, when a node-based field is imported and a surface from the
previous model is specified, the field is extracted from the nodes belonging to the
surface facets of the specified surface. When an element-based field is imported and
a node set from the previous model is specified, the element field is extracted from
elements containing the nodes in the node set. Elements connected to nodes that are
not in the node set are not included.
When importing an external field, you specify the region name space, the region name, and the
physics quantity for the target field and the source field. The region name space
specifies whether the region is a node set, an element set, or a surface. The region
name is the name of a node set, element set, or surface.
When specifying the target field and source field parameters, you should consider the
following conditions:
The source field region name is optional. If it is not specified, the entire
model is used. However, you are encouraged to provide the source field region
name to reduce the computational cost associated with mapping and data transfer.
When defining initial conditions, the target field region name space and physics
quantities are defined by the initial condition definition and, consequently,
are not used.
When defining distributions, the target field region name space, region name,
and physics quantities are defined by the distribution definition and,
consequently, are not used.
When used for node-based submodeling, the target region is a node set.
The physics quantities for the target field and the source field are defined using
identifiers. The physics quantity identifiers for the target field depend on the
analysis procedure. Table 1 lists the supported identifiers. If the source analysis is an
Abaqus analysis, the identifiers are defined in Abaqus/Standard output variable identifiers. Most field-type quantities
can be extracted from the output database file.
Table 1. Supported target field identifiers.
Analysis
Physics Quantity ID
Description
Analysis procedures supporting mechanical degrees of freedom*
U
Displacement
V
Velocity
CF
Force
Analysis procedures supporting thermal degrees of freedom*
NT
Nodal temperature
CFL
Concentrated heat flux at a node
Analysis procedures supporting pore fluid pressure
degree of freedom
POR
Pore pressure
CFF
Concentrated fluid flow at a node
Analysis procedures supporting mass concentration
CFL
Concentrated mass flux at a node
SINV
Equivalent pressure stress
NNC
Normalized concentration
Analysis procedures supporting electric
potential
EPOT
Electric potential
ECD
Electric current density
All procedures*
TEMP
Predefined temperature field
FVn
Predefined field
ESDVn
Element solution-dependent variable
*Procedures excluding explicit dynamics.
Field are mapped automatically when the source and target meshes are topologically different.
However, Abaqus performs a mesh check internally before importing distributions and initial
conditions, and, if the following conditions are satisfied, no mapping is performed:
All of the nodes and elements in the target region are found in the source
output database (.sim) file with matching labels.
Matching elements in the target region and the source
.sim file have the same connectivities and element
types.
Source data are requested at the same location (nodes or element integration
points) where distributions are specified.
Importing Field Variables and Temperature as a Field Variable
When importing field variables and temperature as a field variable, you might notice
differences when you compare the source and target fields. There are two possible
causes of these differences:
Field variables and temperature as a field variable are mapped accurately
and imported at nodes. However, when saved to the output database, these
quantities are interpolated to element integration points. When contouring,
some averaging can occur depending on the contouring parameters
selected.
For fully integrated first-order elements, a selective reduced-integration
technique is employed (see Solid isoparametric quadrilaterals and hexahedra). In this case field variables and
temperature as a field variable are averaged at the element center and saved
to the output database.
The values correctly reflect the internal magnitudes used by Abaqus.
Importing Fields at a Specified Step and Increment
To define distributions or
initial conditions, you can specify the step and the increment from which to extract
the field values. You must ensure that the field is saved to the SIM database at the
specified step and increment. If the increment is omitted, the values at the end of
the step are imported.
Importing Fields at a Specified Time
You can import fields at a specified time. You specify the time to extract field data from the
SIM database. Abaqus linearly interpolates between the time points available in the output database to
obtain field values at the requested time.
You can specify the
analysis step, in which case the time refers to the step time to extract field
values in the output database. If you specify a step without any time parameters,
the last available field values for the analysis step are obtained, which might be
useful when reading from a steady-state analysis where time has no physical
meaning.
Importing Fields over a Specified Time Range
You can import fields over a specified time range when importing fields to define
history-dependent fields. You specify the start and end times to define the time
range in the source analysis to extract field values. Typically, data corresponding
to these time points are not present in the output database. Abaqus linearly interpolates between the time points stored in the output database to
obtain values at the required time. Because the interpolation is linear, you must
ensure that sufficient data is available in the output database to achieve
meaningful interpolation.
You can
specify the analysis step, in which case the time range refers to the analysis step
time to extract field values in the output database.
Resolving Different Physics Time Scales
The physics time scales might be different when performing a sequential analysis. For example,
you can perform an electromagnetic analysis followed by a thermal-stress analysis
importing heat losses. The time scale for the electromagnetic analysis might be
significantly shorter than the time scale for conduction in the thermal-stress
analysis. When the requested time range of the source analysis is different from the
analysis step time of the current analysis, Abaqus automatically scales the time periods.
Applying Field Mapper Controls
Typically, the source and target regions have different meshes. Fields are mapped
conservatively between the source and target meshes, and Abaqus selects the most appropriate mapping algorithm based on the field properties. For
most cases, the default settings for the mapping algorithm are adequate, but you can
modify the mapper settings by specifying field mapper controls when importing
distributions and initial conditions. You cannot specify mapping controls when
importing step-dependent field data or when used for a submodel analysis.
Orphan members might be reported during the mapping operation. Orphan members are nodes or
elements for which the mapper cannot find a correspondence with the source mesh.
Consequently, the mapper does not compute field values at these locations. In this
case, you can modify the default search tolerances or specify how to treat orphan
members.
Applying Field Operations
You can apply one or more field operations when importing fields.
You can scale a field by a constant value by providing a scale factor. The scaling factor is
intended to scale a field by a physical constant. It is not intended for unit
conversion; for example, where you might need additional parameters to convert a
temperature field.
You can extract a component of a vector or tensor field by providing a component index,
i. The ith
component of the source vector or tensor field is extracted and then mapped and
imported as a scalar field. The components are extracted in the local coordinate
system of the source field. You can apply component extraction when importing
distributions, initial conditions, and history-dependent fields; it cannot be used
for a submodel analysis.
If a region in the previous analysis is not colocated with the region in the current
analysis, you must specify the translation and rotation to reposition the source
region before field values are imported. You specify a rotation by giving two
points, a and b, to define a
rotation axis and a right-handed angular rotation about the axis. Local coordinate
systems defined within the source region are translated and rotated according to the
specified positioning data.
Limitations
The following limitations apply:
Volumes or surfaces defined over beam, pipe, and truss elements or edges of
shell or membrane elements cannot be used to specify the target region. When
appropriate, you should import values at nodes.
Source or target regions with the following underlying element types are not
supported: modified triangular elements, modified tetrahedral elements,
second-order temperature-displacement elements, generalized plane strain
elements, and axisymmetric shells.
Because importing external fields in a sequential analysis is a general capability,
no error message is issued if Abaqus encounters the conditions listed above.