计算机科学
短路
工程类
电子工程
实时计算
功率(物理)
电气工程
噪音(视频)
组分(热力学)
信号处理
作者
Jilong Song,Jinshuo Fu,Songtao Che,Hongsi Shi,Kai Wang
标识
DOI:10.1109/aespe65805.2025.11162772
摘要
With the widespread application of lithium-ion batteries in electric vehicles and energy storage systems, safety issues have become increasingly prominent, particularly internal short circuit (ISC) faults, which can easily lead to thermal runaway and serious accidents. This paper proposes an ISC fault diagnosis method based on electrochemical impedance spectroscopy (EIS) data. The method first uses continuous wavelet transform (CWT) to extract multi-scale frequency domain energy features and combines battery state of charge (SOC) and state of health (SOH) to construct multi-dimensional feature vectors. Subsequently, an XGBoost classifier is employed for ISC identification, and Bayesian hyperparameter optimisation is conducted using the Optuna framework. The proposed method was validated using impedance data from 600 NCM811 batteries under simulated short-circuit and normal conditions at different SOH and SOC conditions. Experimental results on 600 impedance samples achieve an accuracy of 94% and recall of 95%, demonstrating the method’s robustness across varying SOH and SOC conditions. The proposed method exhibits excellent classification performance, providing a reliable data-driven solution for early fault warning of lithium-ion batteries.
科研通智能强力驱动
Strongly Powered by AbleSci AI