RESCOMBINE

RESCOMBINE, NUMFILES, Fname, Ext, Lstep, Sbstep, Fact, KIMG, TIME, ANGLE, NSET, ORDER
Reads results from local results files into the database after a distributed memory parallel (Distributed ANSYS) solution.

Compatible Products: – | Pro | Premium | Enterprise | Ent PP | Ent Solver | –

NUMFILES

Number of local results files that are to be read into the database from the distributed memory parallel solution. This number should be equal to the number of processes used in the parallel solution.

Fname

File name (jobname) used during the distributed parallel solution. The file name must be an alphanumeric string (up to 32 characters) enclosed in single quotes.

Ext

File extension for the results files (for example, RST, RTH, RMG, etc.). The file extension must be an alphanumeric string (up to 8 characters) enclosed in single quotes.

Lstep

Load step number of the data set to be read (defaults to 1):

N

 — 

Read load step N.

FIRST

 — 

Read the first data set (Sbstep and TIME are ignored).

LAST

 — 

Read the last data set (Sbstep and TIME are ignored).

NEXT

 — 

Read the next data set (Sbstep and TIME are ignored). If at the last data set, the first data set will be read as the next.

PREVIOUS

 — 

Read the previous data set (Sbstep and TIME are ignored). If at the first data set, the last data set will be read as the previous.

NEAR

 — 

Read the data set nearest to TIME (Sbstep is ignored). If TIME is blank, read the first data set.

LIST

 — 

Scan the results files and list a summary of each load step (KIMG, TIME, ANGLE, NSET, and ORDER are ignored.)

Sbstep

Substep number within Lstep (defaults to the last substep of the load step). For a buckling (ANTYPE,BUCKLE) or modal (ANTYPE,MODAL) analysis, Sbstep corresponds to the mode number (defaults to the first mode). Specify Sbstep = LAST to store the last substep for the specified load step.

If Lstep = LIST, Sbstep = 0 or 1 lists the basic step information; Sbstep = 2 also lists the basic step information, but includes the load step title, and labels imaginary data sets if they exist.

Fact

Scale factor applied to data read from the files. If zero (or blank), a value of 1.0 is used. A nonzero factor excludes non-summable items. Harmonic velocities or accelerations may be calculated from the displacement results from a modal (ANTYPE,MODAL) or harmonic (ANTYPE,HARMIC) analysis. If Fact = VELO, the harmonic velocities (v) are calculated from the displacements (d) at a particular frequency (f) according to the relationship v = 2πfd. Similarly, if Fact = ACEL, the harmonic accelerations (a) are calculated as a = (2πf)2d.

KIMG

Used only with complex results (harmonic and complex modal analyses).

0 or REAL

 — 

Store the real part of a complex solution (default).

1, 2 or IMAG

 — 

Store the imaginary part of a complex solution.

TIME

Time-point identifying the data set to be read. For a harmonic analysis, time corresponds to the frequency. For a buckling analysis, time corresponds to the load factor. Used only in the following cases: If Lstep = NEAR, read the data set nearest to TIME. If both Lstep and Sbstep are zero (or blank), read data set at time = TIME. If TIME is between two solution time points on the results file, a linear interpolation is done between the two data sets. Solution items not written to the results file (OUTRES) for either data set will result in a null item after data set interpolation. If TIME is beyond the last time point on the file, the last time point will be used.

ANGLE

Circumferential location (0.0 to 360°). Defines the circumferential location for the harmonic calculations used when reading from the results file. The harmonic factor (based on the circumferential angle) is applied to the harmonic elements (PLANE25, PLANE75, PLANE78, PLANE83, and SHELL61) of the load case. See the Mechanical APDL Theory Reference for details. Note that factored values of applied constraints and loads will overwrite any values existing in the database.

NSET

Data set number of the data set to be read. If a positive value for NSET is entered, Lstep, Sbstep, KIMG, and TIME are ignored. Available set numbers can be determined by RESCOMBINE,,,,LIST.

ORDER

Key to sort the harmonic index results. This option applies to cyclic symmetry buckling and modal analyses only, and is valid only when Lstep = FIRST, LAST, NEXT, PREVIOUS, NEAR or LIST.

ORDER

 — 

Sort the harmonic index results in ascending order of eigenfrequencies or buckling load multipliers.

(blank)

 — 

No sorting takes place.

Notes

RESCOMBINE is an ANSYS command macro that allows you to combine results from a distributed memory parallel (Distributed ANSYS) solution. Because RESCOMBINE is a command macro, a character string input for any argument must be enclosed in single quotes (for example, 'FIRST' input for Lstep).

In a distributed memory parallel solution, a global results file is saved by default. However, if you issued DMPOPTION,RST,NO in the parallel solution, no global results file is written and all local results files will be kept. In this case, you can use the RESCOMBINE command macro in the general postprocessor (/POST1) to read results into the database for postprocessing.

The RESCOMBINE command cannot be used to combine results from local files generated during a distributed memory parallel solution that used the frequency or cyclic harmonic index domain decomposition method (DDOPTION,FREQ or DDOPTION,CYCHI).

In order to use the RESCOMBINE command, all local results files from the distributed memory parallel solution must be in the current working directory. If running on a single machine, the local results files are saved in the working directory by default. If running on a cluster, the local results files are kept in the working directory on each compute node. For this latter case, you must copy the local results files to the working directory on the primary compute node.

Similar to the SET command, the RESCOMBINE command macro defines the data set to be read from the results files into the database. Various operations may also be performed during the read operation (see the SET command for more details). The database must have the model data available (or use the RESUME command before the RESCOMBINE command to restore the geometry from Jobname.DB).

After a set of data is combined into the database using RESCOMBINE, the RESWRITE command can be used to write this set of data into a new results file. This new results file will essentially contain the current set of results data for the entire (i.e., global) model.

Upon completion of the RESCOMBINE command macro, the current file for postprocessing (see the FILE command) is set to the last local results file specified by RESCOMBINE. For example, if reading in four local results files, the results file for/POST1 will be specified as Jobname3.RST when the RESCOMBINE command macro is complete. Therefore, be aware that some downstream postprocessing actions (such as a SET command) may be operating on only this one local results file.

The RESCOMBINE command macro is intended for use with /POST1. If you want to postprocess distributed parallel solution results using the time-history postprocessor (/POST26), it is recommended that you combine your local results files into one global results file (DMPOPTION,RST,YES).

The load case commands in the general postprocessor (e.g., LCDEF, LCFILE, LCOPER, etc.) are not supported when using RESCOMBINE. These commands set up pointers into the results file used for postprocessing; they cannot be used with the local results files used by RESCOMBINE.

The /CYCEXPAND command used to enable cyclic expansion cannot be used in conjunction with the RESCOMBINE command macro. This is because /CYCEXPAND internally re-reads the results file to do the expansion, and it cannot read multiple local results files.

Menu Paths

This command cannot be accessed from a menu.

Release 18.2 - © ANSYS, Inc. All rights reserved.