AI-Driven Surrogate Models for Predicting Electrode-Scale Discharge Behavior in Lithium-Ion Batteries

2026-07-22

AI Summary

This study addresses the prohibitive computational cost of physics-based simulations at the electrode scale in lithium-ion batteries, which hinders high-throughput design. To overcome this challenge, the authors propose a deep learning surrogate model based on the Swin3D Transformer that directly predicts the spatiotemporal evolution of battery discharge from three-dimensional electrode microstructures. The approach innovatively incorporates Gaussian positional encoding (GPE) to accommodate the complex geometry of porous electrodes and introduces a dedicated temporal encoding module to capture nonlinear electrochemical dynamics. Experimental results demonstrate that the proposed method significantly outperforms existing 3D point-cloud-based baselines on an electrochemical simulation dataset, achieving high prediction accuracy while reducing computational cost by several orders of magnitude.

Abstract

Physics-based simulations are essential for understanding the electrode-scale discharge behavior of lithium-ion batteries (LIBs) but suffer from prohibitive computational costs. To address this, we introduce a novel deep learning surrogate pipeline based on the Swin3D Transformer to predict spatiotemporal discharge dynamics directly from volumetric data. Our approach integrates two key innovations: Gaussian Positional Encoding (GPE), which enhances spatial feature representation by adapting to the complex geometry of electrode microstructures, and a specialized Temporal Encoding module to capture non-linear timeseries evolution. Experimental validation on an Electrochemical Simulation (ES) dataset demonstrates that our pipeline significantly outperforms state-of-the-art point cloud baselines in prediction accuracy. Furthermore, the proposed method reduces the computational overhead by orders of magnitude, providing a scalable and efficient framework for high-throughput battery design and optimization.

Problem

• lithium-ion batteries
• electrode-scale discharge behavior
• computational cost
• surrogate modeling
• spatiotemporal dynamics

Innovation

• Swin3D Transformer
• Gaussian Positional Encoding
• Temporal Encoding
• surrogate modeling
• electrode-scale discharge prediction