Fusing Deep Neural Networks with Multi-PDE Solvers: An Enhanced Approach to Battery Modeling

Qinshuo Liu, Lishuo LIU, Jianfeng Zhang, Lujia Pan, Guodong Li

Keywords: AI for battery, P2D model, Pinns model, parameter estimation

TL;DR: Neural networks solutions for P2D model in battery

Abstract: Accurately characterizing the migration behaviors of microscopic particles within lithium batteries is of great significance for the design and development of new battery technologies. During charge and discharge cycles, the intercalation and deintercalation of lithium ions between the cathode and anode involve complex multi-physics coupling processes. Traditional numerical methods often suffer from low computational efficiency when simulating such dynamic systems. To address this challenge, we propose a novel deep learning framework that directly incorporates the dynamic evolution of the system into the neural network architecture. Our approach ensures reliable parameter estimation and voltage reconstruction while significantly improving computational inference efficiency. Extensive simulation experiments validate the effectiveness of the proposed method, offering a new technical pathway for high-precision and efficient battery modeling.

Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)

Submission Number: 3098