Example proposal: link `bndout` output with input `deck` (node) sets

[Jalmenara]

📝 Description of the example

Node sets in bndout

As a user, I needed to learn the relation between a given input deck (in particular, its node sets) and the output present in the bndout ascii file, accessed through LS Pre Post.

For that, I created with Copilot a "mini, boilerplate" python API which helped me be certain about it ( "lsdyna_bndout_learner.zip" attached below). The set1_vs_set2_comparison_analysis.py summarizes the use case.

The value it brings to the user is relating the node set entities in the input deck (*SET_NODE_LIST kwds.) with the contents of the bndout file. A link which doesn't exist out of the box (afaik, only the nodes are present there; any insights about it are welcomed :) ). In fact, if the bndout is imported to LS-PrePost as ascii file, the node sets don't appear there either (screenshot below)

*SET_NODE_LIST
         1
      4801      4802      4803      4804      4805      4806      4807      4808
      4809      4810      4811      4812      4813      4814      4815      4816
      4817      4818      4819      4820      4821      4822      4823      4824

                                                                         
                    ls-dyna smp s R16 R16.1.1_20_0c90ca date 10/01/2025

 n o d a l   f o r c e/e n e r g y    o u t p u t  t=   1.89373E-06

 discrete nodal point forces 

 nd#    4681  xforce=   0.0000E+00   yforce=   0.0000E+00  zforce=   0.0000E+00   energy=   0.0000E+00
 nd#    4682  xforce=   0.0000E+00   yforce=   0.0000E+00  zforce=   0.0000E+00   energy=   0.0000E+00
 nd#    4683  xforce=   0.0000E+00   yforce=   0.0000E+00  zforce=   0.0000E+00   energy=   0.0000E+00
 nd#    4684  xforce=   0.0000E+00   yforce=   0.0000E+00  zforce=   0.0000E+00   energy=   0.0000E+00
 nd#    4685  xforce=   0.0000E+00   yforce=   0.0000E+00  zforce=   0.0000E+00   energy=   0.0000E+00
 nd#    4686  xforce=   0.0000E+00   yforce=   0.0000E+00  zforce=   0.0000E+00   energy=   0.0000E+00
 nd#    4687  xforce=   0.0000E+00   yforce=   0.0000E+00  zforce=   0.0000E+00   energy=   0.0000E+00

Bringing that need to PyDyna

For instance, in PyDyna I could only find deck.get_kwds_by_type, whereas in the boilerplate API, there was a get_node_set_by_id available.

node_sets = list(deck.get_kwds_by_type("SET")) ## pydyna
node_set = parser.get_node_set_by_id(set_id) ## lsdyna_parser

I'd like to incorporate some of the boilerplate API concepts to PyDyna, as they can be useful for somebody else. For that, I used AI to create a _pydyna.py flavour of the script, which outputs the same results. In it, I missed mainly two features

• a) The concept of "set". I think at least a refinement of the class Deck could work for that. See feat: add Deck.get_set_by_id() and extend get_kwds_by_full_type() #1066
• b) Linking "sets" to concepts like "boundary conditions", as per Add ability for keywords to point to each other #599
• c) Loading and handling of the bndout file and its contents.

b) and c) area surely more complex tasks, and I am basically using this issue to jumpstart my contributions to this repo, in alignment with them. Please don't hesitate to raise any concerns regarding these ideas.
Specifically, I believe I must check the PyDPF-Post project, because the necessary contributions will likely belong there. Also, I'd like to explore bringing some of the concepts in PyMechanical as well.

📁 Files needed for running the example

lsdyna_bndout_learner.zip

pip install -r requirements.txt

Or install manually:

pip install pandas numpy matplotlib ipykernel pytest pyyaml

Files

.
├── lsdyna_parser.py              # Input file parsing module
├── bndout_parser.py              # Results file parsing module  
├── lsdyna_postprocessor.py       # Integrated post-processor
├── example_postprocess.py        # Example standalone script
├── example_bndout_analysis.ipynb # Interactive notebook
├── requirements.txt              # Python dependencies
└── README.md                     # This file

Usage

Quick Start - Complete Example

The easiest way to get started is to run the complete example script:

python set1_vs_set2_comparison_analysis.py

...

💻 Which operating system are you using?

Windows

📀 Which ANSYS version are you using?

ANSYS 2025 R2, Ansys LS-DYNA, LS-PrePost

🐍 Which Python version are you using?

3.10

📦 Installed packages

pandas 
numpy 
matplotlib
ansys-dyna-core

[koubaa]

I think it would be acceptable for pydyna to have it's own small post-processing library, especially to read text files. We can even implement some dpf operators around this and contribute them to pydpf as a release artifact.