计算机科学
电池(电)
估计
人工神经网络
实时计算
噪音(视频)
人工智能
估计理论
工程类
算法
控制理论(社会学)
作者
Zekariyas Yohannes Abebe,Yonghong Xu,Yidi Wei,Jian Li,Boru Jia
出处
期刊:Energy and AI
[Elsevier BV]
日期:2026-10-05
卷期号:26: 100928-100928
标识
DOI:10.1016/j.egyai.2026.100928
摘要
ABSTRACT Accurate, robust state-of-health (SOH) estimation is essential for safe battery operation and second-life deployment, yet most existing methods depend on fixed feature extraction windows rarely available in real-world charging and offer limited interpretability or cross-chemistry generalization. To address these issues, we propose the physics-informed efficient Kolmogorov-Arnold network (PI-EKAN), which estimates SOH from adaptive partial charging. Prefix-specific Bayesian optimization (PS-BO) selects the most informative state-of-charge window for each chemistry and extracts seven voltage-based descriptors; this feed-coupled solution and dynamics networks are trained under a bounded uncertainty-weighted objective that keeps the data, physics, and monotonicity constraints balanced. Across eight datasets, PI-EKAN reaches an average MAPE of 0.466%, an RMSE of 0.529%, and an R² above 0.979, cutting MAPE by 35.6% relative to a physics-informed baseline while using only 1,800 parameters, far fewer than comparable recurrent Kolmogorov-Arnold models. PS-BO identifies chemistry-specific charging windows that stay informative under restricted SOC visibility, reduced sampling frequency, and simulated battery-management-system errors. In cross-chemistry tests, adapting the pretrained model with one labeled target cell cuts average MAPE from 6.35% source-only transfer to 1.19% using hybrid low-rank EKAN-LoRA fine-tuning with half the trainable parameters across sixteen transfer pairs. Component ablation identifies the bounded adaptive weighting as the key stabilizing element, and post-hoc attribution and learned-spline analyses consistently rank voltage-shape descriptors as the leading inputs, indicating that the model captures degradation-related structure rather than dataset artifacts. These results establish PI-EKAN as a compact, interpretable, chemistry-transferable framework for SOH estimation from the incomplete charging data real battery systems produce.
科研通智能强力驱动
Strongly Powered by AbleSci AI