This repository contains the cached fields needed to reproduce the WiSED
controlled PDE benchmarks and engineering validations. The associated source
code is maintained in
Hust-WiSED.
| Directory | Contents | Provenance |
|---|---|---|
pde_benchmarks/ |
28 noisy and noise-free fields for four controlled PDE benchmarks across seven noise levels. | Generated by numerical solvers in Hust-WiSED. |
sib_diffusion/ |
Generated radial concentration field for the retained 3% noise SIB experiment. | Generated by the SIB simulation in Hust-WiSED. |
soc_eis/ |
Public raw EIS and capacity measurements for batteries B01-B11. | Mustafa et al., Mendeley Data V2. |
The pde_benchmarks/ archives use Git LFS. Clone both repositories, retrieve
the LFS objects, then run this command from the Hust-WiSED root:
cd ../Hust-WiSED-Dataset
git lfs install
git lfs pull
cd ../Hust-WiSED
python scripts/install_release_data.py --dataset-root ../Hust-WiSED-DatasetThe installer copies the benchmark archives into data/dataset/, the SIB field
into the same directory, and the SOC-EIS measurements into the engineering
case directory. It does not alter this dataset repository.
See pde_benchmarks/README.md for benchmark
equations, shapes, noise protocol and archive schema.
The raw soc_eis/ directory preserves the original B01-B11 battery and test
cycle layout. It was obtained from:
Mustafa, H., Bourelly, C., Vitelli, M., Milano, F., Molinara, M. and Ferrigno, L. SoC Estimation on Li-ion Batteries: A New EIS-based Dataset for data-driven applications. Mendeley Data, V2 (2024). DOI: 10.17632/cb887gkmxw.2.
The associated data article is Mustafa et al., Data in Brief 57, 110947 (2024). DOI: 10.1016/j.dib.2024.110947. Please cite both records and retain the original CC BY 4.0 attribution when reusing any SOC-EIS measurement or derivative result.
See LICENSES.md for the component-specific terms. In particular, the public
SOC-EIS measurements remain subject to the original authors' CC BY 4.0 license;
this repository does not replace their attribution or licensing conditions.
Each generated-data directory contains a README describing its schema, provenance and reuse conditions. Git LFS records a SHA-256 object identifier for every benchmark archive.