Autonomous End-to-End SOH Prediction Services for Battery Systems via Temporal-Contrastive Representation Learning

2026-06-15

AI Summary

This work addresses the challenges in industrial deployment of lithium-ion battery state-of-health (SOH) estimation, which often stems from reliance on manual feature engineering and opaque black-box models. To overcome these limitations, the authors propose TC-SOH, a modular and plug-and-play framework that leverages temporal contrastive learning and cross-window prediction as self-supervised pretraining tasks to learn degradation-aware representations directly from raw operational data in an end-to-end manner. Through integrated visualization, sensitivity analysis, and multiple probing techniques, the study demonstrates that the learned features not only encompass established expert-derived indicators but also capture additional SOH-relevant information, highlighting the critical role of temporal structure in accurate prediction. Evaluated on four public datasets, TC-SOH substantially outperforms existing physics-informed and data-driven methods, achieving 1.91× and 2.13× reductions in MAPE and RMSE, respectively.

Abstract

Accurate state of health (SOH) estimation is a critical diagnostic service for lithium-ion battery management. However, reliance on labor-intensive manual feature engineering and opaque black-box models hinders scalable industrial deployment. To address this, we introduce TC-SOH: a modular, plug-and-play service architecture for autonomous, end-to-end SOH prediction. TC-SOH employs a temporal-contrastive mechanism and a cross-window prediction pretext task to extract degradation-relevant representations directly from raw operational data. To improve transparency, we connect model efficacy with representation diagnostics: visualization, sensitivity analysis, redundancy analysis, bidirectional probing, future-SOH probing, and temporal shuffling show that learned features overlap with selected expert descriptors while retaining additional SOH-relevant variation, and that ordered temporal context improves subsequent-SOH prediction. Across four public datasets, TC-SOH outperforms the considered physics-informed and data-driven baselines, reducing MAPE by 1.91 times and RMSE by 2.13 times.

Problem

• state of health
• battery management
• feature engineering
• black-box models
• scalable deployment

Innovation

• temporal-contrastive learning
• end-to-end SOH prediction
• representation learning
• battery state of health
• interpretable AI