Comparative Study of Neural Surrogate Architectures for Autoregressive Prediction of Internal Battery States

2026-06-18

AI Summary

This study addresses the challenge of deploying high-fidelity electrochemical battery models—such as the Doyle–Fuller–Newman (DFN) model—in real-time battery management systems due to their prohibitive computational cost, while existing machine learning surrogates often suffer from limited generalization. Within a unified framework, the authors systematically compare four neural architectures—MLP, ResNet, U-Net, and Fourier Neural Operator (FNO)—as autoregressive state-transition operators, for the first time isolating the impact of architectural inductive biases on surrogate performance under controlled conditions. Their approach integrates multi-step unrolling, current-condition embedding, and GPU acceleration. Experimental results demonstrate that U-Net achieves superior generalization, yielding an average normalized root-mean-square error of only 3% over 300-step autoregressive predictions and accelerating inference by 5.38× compared to conventional numerical solvers.

Abstract

The Doyle-Fuller-Newman (DFN) model resolves internal electrochemical states in lithium-ion batteries with high fidelity. However, the numerical solution of its governing equations is computationally prohibitive for real-time deployment, limiting scalability from individual cells to pack and fleet-scale applications. While machine learning surrogates can substantially reduce inference latency through GPU acceleration, most existing approaches learn solution approximations tied to specific operating conditions rather than learning generalizable state-evolution dynamics. This work presents a systematic comparison of four neural network architectures (MLP, ResNet, U-Net, FNO) formulated as autoregressive state-transition operators that predict full DFN internal states across a wide range of operating conditions. To ensure a controlled architectural comparison, all models are trained under a unified framework using multi-step unrolling and current-conditioning, isolating the impact of spatial inductive bias. Results demonstrate that the U-Net's multi-scale feature hierarchy achieves a mean final-step nRMSE of 3% averaged across all internal state variables after 300-step autoregressive rollouts, while providing a 5.38x speed-up over the numerical solver. These findings highlight spatial inductive bias as a critical determinant of surrogate performance, advancing the development of surrogates for internal state observability for next-generation battery management systems and digital twins.

Problem

• neural surrogate
• autoregressive prediction
• battery internal states
• generalizable dynamics
• real-time deployment

Innovation

• neural surrogate
• autoregressive prediction
• spatial inductive bias
• U-Net
• battery state estimation