Empowering lithium-ion battery state of health solution: Incorporating electrochemical aging characterization and unscented particle filter

Accurate evaluation of the State of Health (SOH) is crucial for the safe and effective integration of Lithium-Ion Batteries (LIBs) in Electric Vehicles and energy storage systems. This study proposes a battery SOH estimation method incorporating electrochemical aging characterization and an Unscented Particle Filter (UPF) to address the lack of physical mechanisms in conventional estimation. First, the Pseudo-two-dimensional (P2D) electrochemical model and its governing equations are employed to establish the internal physical foundation the battery. Structural parameters are determined via disassembly experiments, while performance parameters are obtained by coupling the P2D model with the Genetic Algorithm (GA). The stoichiometric ratio from the P2D model is used to deduce capacity, forming the observation equation. A double exponential model is then established as the state equation to describe capacity degradation, formulating a mathematical state-space representation. Finally, the UPF method is applied to effectively track the SOH degradation and correct the state-space representation for next estimation. Aging experiment results demonstrate that the proposed solution exhibits superior accuracy in terms of Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) compared to the conventional P2D model and the pure data-driven methods, such as Gaussian Process Regression (GPR) and Support Vector Regression (SVR). The framework for battery SOH estimation incorporating Electrochemical Aging Characterization and Unscented Particle Filter. • Novel P2D-UPF fusion for physics-informed lithium-ion battery SOH estimation. • P2D model + disassembly experiments derive electrochemical parameters via GA. • Battery aging State-space model links degradation dynamics. • Validated framework enhances SOH accuracy for battery safety and reliability.