• How does NESSUS interface to MSC/NASTRAN?
• How can I take advantage of the batch/restart capabilities?
• What is the density function for the maximum entropy distribution?
• How can I execute NESSUS from the command line?
• Where may I direct questions concerning the usage of NESSUS?

How does NESSUS interface to MSC/NASTRAN??

NESSUS has a full featured interface to MSC/NASTRAN. However, a few requirements are listed below that allow NESSUS to effectively communicate with MSC/NASTRAN.

MSC/NASTRAN Input Card Requirements

One card is required for in the MSC/NASTRAN data input to properly write the output2 file for use with NESSUS. The card instructs MSC/NASTRAN to write all results to the output2 file. This card must be placed in the case control section of the input.

MSC/NASTRAN Execution Parameters

NESSUS requires multiple NASTRAN executions during a probabilistic analysis. By default, NASTRAN is executed with the default execution parameters and is submitted as a background process. Additional execution parameters are required to prevent the NASTRAN run from being executed in the background. At a minimum, the bat=no option is required after the command string. Additional execution arguments are recommended to reduce unneeded files. These are scr=yes old=no news=no which instructs NASTRAN to remove the scratch files, do not save old files, and do not print the news to the output file. The GUI automatically defines the recommended execution command. Other command line arguments can be included if needed. For example, other arguments may be required to increase memory or change the scratch directory for the NASTRAN analyses.

MSC/NASTRAN Response Selection

Displacements, accelerations, and frequencies are selected directly in the GUI.

Stress and strain components are selected based on the NASTRAN element stress/strain item codes (MSC/NASTRAN Quick Reference Guide, Appendix A). Listed below are the stress/strain item codes for the QUAD4 element. As an example, to choose the von Mises stress at Z1 for a QUAD4 element, the component number would be 9.

Page A-11, Appendix A of the "MSC/NASTRAN Quick Reference Guide," Version 68, Michael Reymond and Mark Miller, Editors, The MacNeal-Schwendler Corporation, February 1994.

How can I take advantage of the batch/restart capabilities?

Restarting NESSUS is fully automatic when all/part of the performance function is defined with numerical models. During a NESSUS analysis, a database (jobname.rdb) is created in a directory named for the model that contains the random variable values and the results file name. For a given a set of random variables, NESSUS attempts to match the set in the restart database. If there is a match, then NESSUS reuses the results file if it exists and does not run the analysis code. If the file does not exist, an input file is created with the set of random variables, a new entry is added to the restart database and the analysis code is run. Thus, NESSUS can be restarted at any point in the probabilistic analysis by simply resubmitting the original job.

If the batch option is used, NESSUS will only create the input files where the results do not exist.

Some capabilities and/or ways the restart capability can be used are:

1. Recover from a computer crash that occurred during the MV or RSM perturbation analysis. The finite element perturbation analyses will begin automatically at the point where the run crashed.
2. For large jobs, the MV or RSM analyses can consume considerable time. Using the restart option, the MV analysis can be examined, and if desired, an AMV or AMV+ analysis started without having to repeat the MV analysis.
3. Once the perturbation files are created, looping over multiple nodes and/or elements is very fast since the perturbations files are not re-created.

What is the density function for the maximum entropy distribution?

The probability density function for the maximum entropy distribution is defined as

The lower and upper bounds of X must also be specified. The lower limit for k is 1.

How can I execute NESSUS from the command line?

Command line execution format for Unix and Linux systems:

~/nessus/bin/SunOS/nessus.exe ~/nessus/license/license.txt jobname [deterministic|variation|probabilistic] probabilistic [quiet|verbose|noisy] [cvar]

The SunOS in the path above is system dependent. List the contents of the nessus/bin directory to find the name for your system.

Command line execution format for Windows systems using the default installation directory:

C:\nessus\bin\w32\nessus.exe C:\nessus\license\license.txt jobname [deterministic|variation|probabilistic] probabilistic [quiet|verbose|noisy] [cvar]

Arguments:

One option in the brackets [ ] is required except for cvar. The brackets are not included in the actual command.

[cvar] is only needed for deterministic and variation analysis and is the name of the response variable of interest.

The ~ represents a relative path to the user's home area and the forward slash "/" syntax is for the Unix and Linux environments. Full path information may need to be provided to point to the NESSUS installation directory

[deterministic|variation|probabilistic] = selected analysis mode

[quiet|verbose|noisy] = print verbosity mode

jobname = NESSUS input file without the .dat extension. A path is required if the jobname.dat file is not located in the current working directory.

Where may I direct questions concerning the usage of NESSUS?

Please email any questions about NESSUS to nessus@swri.org