Official implementation of the paper "Lightweight Physics-Informed Reservoir Computing for Battery Health Prediction", accepted at IFAC MECC 2026.

PIRC is a lightweight prognostic framework that integrates physical degradation priors directly into a closed-form ridge regression readout. It employs an adaptive switching mechanism to autonomously select the optimal feature representation, mitigating over-parameterization while strictly minimizing trainable parameters.

If you use this code in your research, please cite our paper:

@inproceedings{anurag2026pirc,
  title={Lightweight Physics-Informed Reservoir Computing for Battery Health Prediction},
  author={Anurag, Kumar and Xu, Yanwen and Wan, Wenbin},
  year={2026}
}

The framework relies solely on standard scientific Python libraries. No heavy deep learning frameworks are required.

pip install numpy scipy pandas matplotlib

Official dataset link: https://www.nasa.gov/intelligent-systems-division/discovery-and-systems-health/pcoe/pcoe-data-set-repository/

1. Battery SOH Estimation (NASA PCoE Dataset)

python run_battery.py

2. Chaotic Systems (Lorenz & Rössler)

python run_chaotic.py

• run_battery.py – Main execution script for the battery SOH benchmark.
• run_chaotic.py – Main execution script for the chaotic systems benchmark.
• pirc_core.py – Core implementations of the proposed PIRC and Hybrid RC-NGRC models.
• reservoir.py – Reservoir computing utilities, nonlinear transforms, and NVAR feature construction.
• baselines.py – Implementations of baseline models (Standard RC and PINN).
• systems.py – Dynamical system definitions (Lorenz, Rössler, Mackey-Glass).
• metrics.py – Evaluation metrics and plotting utilities.
• 5_Battery_Data_Set/ – Directory containing the NASA PCoE dataset.

Kumar Anurag