A hybrid physics-artificial intelligence framework that integrates a Large Language Model as a high-level behavioral reasoning layer within a physics-driven supervisory architecture provides a principled pathway for safe AI integration in EV energy management.
Abstract
Electric vehicle (EV) battery performance is strongly coupled with driver behavior, yet human intent is typically expressed semantically rather than numerically. This paper proposes a hybrid physics-artificial intelligence framework that integrates a Large Language Model (LLM) as a high-level behavioral reasoning layer within a physics-driven supervisory architecture. The LLM interprets textual user intent and structured battery feedback to generate bounded behavioral parameters that shape a discharge current envelope. A physics-driven safety filter then enforces physical safety constraints before computing feasible velocity recommendations. Lyapunov-based analysis establishes bounded recommendation error under battery model and prompt inaccuracies. Simulation results demonstrate adaptive, user-aware operation without compromising physical safety. The proposed reasoning-enforcement architecture provides a principled pathway for safe AI integration in EV energy management.
The State of Health (SOH) estimation of the lithium-ion batteries is essential for safe and efficient electric vehicle operation. Existing techniques often used black-box models, needing interpretability and the interaction of the user. This paper represents a multi-paradigm AI framework that combines deep learning, ne...
Ghada Toumi, A. Sahbani, H. Mahersia· International Symposium on N...· 0 citations
Energy infrastructure systems, e.g. EV charging networks, smart buildings, must make real-time decisions to satisfy time-varying, context-dependent human demand. Yet existing methods fall short: edge controllers cannot interpret human-demand contexts, whereas LLM agents that excel at fusing such human information are u...
Tong Zhou, Jia-Qi Wen, Jian-Yi Yang· Proceedings of the 4th Inter...· 0 citations
Accurate battery models must generalize to unseen operating routes while supporting terminal-voltage prediction and recursive state-of-charge (SOC) estimation. This study evaluates whether physics-guided residual learning improves complete-route generalization compared with increasing equivalent-circuit-model (ECM) ord...
J. Valladolid, Juan P. Ortiz, G. Gruosso et al.· Batteries· 0 citations
It was shown that the proposed PINN outperforms traditional neural networks in terms of accuracy (mean absolute errors), and the use of an attention mechanism emphasizes the role played by recent driving behavior in improving the predictive performance.
K. R. Jeevakamal, Rayappa David Amar Raj, Archana Pallakonda et al.· Neural computing & applicati...· 0 citations
Battery management systems (BMSs) are increasingly required to convert incomplete electrical and thermal measurements into safe control actions rather than merely report battery states. This Mini Review focuses on two deployment questions that connect electrochemistry with control: how electrochemical states can be mad...
Feng Li, Yuan Li, Wen-Qi Wu et al.· Frontiers in Chemistry· 0 citations
The proposed FLY-EVAL++ is an evidence-driven evaluation protocol that combines deterministic verification of protocol compliance, physical feasibility, and safety constraints with fixed rubric-guided aggregation into interpretable multi-dimensional scores.
Ya-Lun Wu, Jun-Feng Fang, Jia-Wei Wang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
