Deep Drawing and Cutting Simulations (DDACS) Dataset - Teaser
Description
A 22 MB sample of the DDACS dataset: one full FEM simulation of a two-stage sheet metal stamping process (deep drawing in OP10, trimming with elastic recovery in OP20), the Croissant 1.1 manifest, the process parameter table for all 32,466 simulations, and the dataset documentation PDF.
This record is a discovery sample. The full release (32,466 simulations, ~640 GB of lossless HDF5, predefined 25,973/3,246/3,247 train/val/test split) is hosted on DaRUS. Please cite the dataset there: doi:10.18419/DARUS-4801.
Quickstart: pip install ddacs · Docs: ddacs.readthedocs.io · Source: github.com/BaumSebastian/DDACS
License: data CC BY 4.0, package code MIT.
Layout note: to use this bundle with the ddacs package, place 258864.zip under data/h5/ next to process_parameters.csv and the manifest. This is the layout that ddacs download --small creates. Zenodo stores files flat, so the folder cannot be preserved here.
Methods
All 32,466 simulations were run with LS-DYNA, modeling the two-stage stamping process of a parameterized cup geometry: deep drawing (OP10) followed by trimming with elastic recovery (OP20). Eight input parameters (four geometric, four process) are sampled across the design space; each simulation stores mesh, displacement, stress, strain, and thickness fields per operation and timestep as HDF5. The Croissant 1.1 manifest in this record declares the complete schema.
Files
ddacs_documentation.pdf
Additional details
Related works
- Is documented by
- Software documentation: https://ddacs.readthedocs.io (URL)
- Is supplement to
- Journal article: 10.1051/matecconf/202540801090 (DOI)
Dates
- Collected
-
2024-10-01/2024-10-31Simulation campaign
Software
- Repository URL
- https://github.com/BaumSebastian/DDACS
- Programming language
- Python
- Development Status
- Active