GPINND: A deep-learning-based state of health estimation for Lithium-ion battery

TL;DR

This paper introduces a deep learning framework integrated with electrochemical models for real-time lithium-ion battery health estimation, achieving high accuracy with reduced computational cost.

Contribution

It presents a hybrid-driven surrogate model and a sequential training strategy for efficient, non-iterative SOH estimation combining physics-based and data-driven methods.

Findings

Voltage reconstruction RMSE of 0.0198 V

SOH estimation RMSE of 0.0014

Effective mitigation of convergence issues

Abstract

Electrochemical models offer superior interpretability and reliability for battery degradation diagnosis. However, the high computational cost of iterative parameter identification severely hinders the practical implementation of electrochemically informed state of health (SOH) estimation in real-time systems. To address this challenge, this paper proposes an SOH estimation method that integrates deep learning with electrochemical mechanisms and adopts a sequential training strategy. First, we construct a hybrid-driven surrogate model to learn internal electrochemical dynamics by fusing high-fidelity simulation data with physical constraints. This model subsequently serves as an accurate and differentiable physical kernel for voltage reconstruction. Then, we develop a self-supervised framework to train a parameter identification network by minimizing the voltage reconstruction error. The resulting model enables the non-iterative identification of aging parameters from external measurements. Finally, utilizing the identified parameters as physicochemical health indicators, we establish a high-precision SOH estimation network that leverages data-driven residual correction to compensate for identification deviations. Crucially, a sequential training strategy is applied across these modules to effectively mitigate convergence issues and improve the accuracy of each module. Experimental results demonstrate that the proposed method achieves an average voltage reconstruction root mean square error (RMSE) of 0.0198 V and an SOH estimation RMSE of 0.0014.