Pace: Physics-Aware Attentive Temporal Convolutional Network for Battery Health Estimation

TL;DR

Pace is a physics-aware attentive temporal convolutional network that accurately estimates battery health by integrating sensor data with physics features, outperforming existing models and suitable for real-time deployment.

Contribution

The paper introduces Pace, a novel neural network architecture combining physics-based features with attention mechanisms for improved battery health estimation.

Findings

Pace outperforms existing models with 6.5 and 2.0x performance improvements.

Achieves high accuracy in diverse battery usage conditions.

Demonstrates real-time deployment on Raspberry Pi.

Abstract

Batteries are critical components in modern energy systems such as electric vehicles and power grid energy storage. Effective battery health management is essential for battery system safety, cost-efficiency, and sustainability. In this paper, we propose Pace, a physics-aware attentive temporal convolutional network for battery health estimation. Pace integrates raw sensor measurements with battery physics features derived from the equivalent circuit model. We develop three battery-specific modules, including dilated temporal blocks for efficient temporal encoding, chunked attention blocks for context modeling, and a dual-head output block for fusing short- and long-term battery degradation patterns. Together, the modules enable Pace to predict battery health accurately and efficiently in various battery usage conditions. In a large public dataset, Pace performs much better than existing models, achieving an average performance improvement of 6.5 and 2.0x compared to two best-performing baseline models. We further demonstrate its practical viability with a real-time edge deployment on a Raspberry Pi. These results establish Pace as a practical and high-performance solution for battery health analytics.