Synthesizing physics and quantum machine learning for next-generation battery health prognostics: A critical review and roadmap

Le résumé fourni par la source

This review systematizes the field of battery health prediction by introducing a novel classification framework for State-of-Health (SOH) and Remaining Useful Life (RUL) models. The landscape is categorized into three distinct generations: purely data-driven machine learning (Gen. 1), Physics-Informed Machine Learning (PI-ML, Gen. 2), and the emerging Quantum Machine Learning (QML, Gen. 3). Based on this taxonomy, a critical benchmark of representative models is presented, evaluating the predictive accuracy across recent AI models to guide practitioners in model selection. Finally, limitations of current approaches are identified, and a forward-looking research roadmap is proposed for Physics-Informed Quantum Machine Learning (PI-QML) as the next evolutionary stage. This roadmap details strategies to address persistent challenges in battery health prognostics. By establishing a clear framework, providing a comparative analysis, and charting a path for future research, this work provides a structured agenda to accelerate the development of next-generation battery health management systems.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.