Unsupervised learning early warning of lithium battery failure driven by cloud data

doi:10.3969/j.issn.1674-8484.2025.02.006

Abstract

Abstract:

An unsupervised learning early warning method was proposed based on voltage consistency to warn early the lithium battery faults in cloud battery management technology. The voltage characteristics in the effective charging cycle were extracted with measuring the degree of voltage consistency by using a minimum neighborhood radius which achieved a single cluster number for DBSCAN (density-based spatial clustering of applications with noise); A parameter with dimension-one was defined to improve the algorithm generalization ability to the actual working conditions; The hyperparameters such as alarm thresholds were selected through orthogonal experiment. The actual fault cases were verified and analyzed. The results show that for the battery systems with the low state of charge (SOC) faults, the single battery undervoltage faults, and the single consistency faults, this method enables early warning more than 50 days in advance, with an accuracy rate of 96.7%, and can locate the cells of subsequently develops faults. Therefore, early warning of lithium-battery-system failures is realized through unsupervised learning.

Key words: electric vehicle, lithium-ion battery (LiB), battery management, cloud data, unsupervised learning, fault warning, minimum neighborhood radius

CLC Number:

U46
TM912

ZHOU Zhengyi, YANG Lin, MENG Yizhen, LI Huaijin, LÜ Feng, LIU Zhisheng, LI Yang, WU Weikun. Unsupervised learning early warning of lithium battery failure driven by cloud data[J]. Journal of Automotive Safety and Energy, 2025, 16(2): 234-242.

Figures/Tables 14

References 23

[1]	Waseem M, Ahmad M, Parveen A, et al. Battery technologies and functionality of battery management system for EVs: Current status, key challenges, and future prospectives[J]. J Power Sources, 2023, 580: Paper No 233349.
[2]	XIONG Rui, SUN Wanzhou, YU Quanqing, et al. Research progress, challenges and prospects of fault diagnosis on battery system of electric vehicles[J]. Appl Energ, 2020, 279: Paper No 115855.
[3]	Schmid M, Gebauer E, Hanzl C, et al. Active model-based fault diagnosis in reconfigurable battery systems[J]. IEEE Trans Power Elect, 2021, 36(3): 2584-2597.
[4]	YAO Li, XIAO Yu, GONG Xiao, et al. A novel intelligent method for fault diagnosis of electric vehicle battery system based on wavelet neural network[J]. J Power Sources, 2020, 453: Paper No 227870.
[5]	WANG Zhenbo, SONG Chunbao, ZHANG Lei, et al. A data-driven method for battery charging capacity abnormality diagnosis in electric vehicle applications[J]. IEEE Trans Transport Electrif, 2022, 8(1): 990-999.
[6]	XIONG Rui, YANG Rui, CHEN Zhiqiang, et al. Online fault diagnosis of external short circuit for lithium-ion battery pack[J]. IEEE Trans Ind Elect, 2020, 67(2): 1081-1091.
[7]	YANG Ruixin, XIONG Rui, HE Hongwen, et al. A fractional-order model-based battery external short circuit fault diagnosis approach for all-climate electric vehicles application[J]. J Clean Produ, 2018, 187: 950-959.
[8]	YAO Lei, ZHENG Jie, XIAO Yanqiu, et al. An intelligent fault diagnosis method for lithium-ion battery pack based on empirical mode decomposition and convolutional neural network[J]. J Energ Stor, 2023, 72: Paper No 108181.
[9]	ZHANG Chun, WANG Qiyue, JIANG Jiuchun, et al. Voltage fault diagnosis of a power battery based on wavelet time-frequency diagram[J]. Energy, 2023, 278: Paper No 127920.
[10]	SUN Zhenyu, WANG Zhenbo, LIU Peng, et al. An online data-driven fault diagnosis and thermal runaway early warning for electric vehicle batteries[J]. IEEE Trans Power Elect, 2022, 37(10): 12636-12646.
[11]	LI Da, ZHANG Zhaosheng, LIU Peng, et al. Battery fault diagnosis for electric vehicles based on voltage abnormality by combining the long short-term memory neural network and the equivalent circuit model[J]. IEEE Tran Power Elect, 2021, 36(2): 1303-1315.
[12]	刘启全, 马建, 赵轩, 等. 基于优化熵算法的真实场景下电动汽车动力电池系统故障诊断[J]. 中国公路学报, 2024, 37(10): 233-248.
	LIU Qiquan, MA Jian, ZHAO Xuan, et al. Fault diagnosis of electric vehicle power battery System based on optimal entropy algorithm in Real scenario[J]. China J Highway Transport, 2024, 37(10): 233-248. (in Chinese)
[13]	金浩博, 王淑青, 汤璐, 等. 面向实际工况的动力电池早期故障诊断研究[J]. 昆明理工大学学报(自然科学版), 2024, 49(2): 73-80.
	JIN Haobo, WANG Shuqing, TANG Lu, et al. Research on early fault diagnosis of power battery under actual conditions[J]. J Kunming Univ of Technology (Nat Sci Ed), 2024, 49(2): 73-80. (in Chinese)
[14]	喻思维, 张雪松, 林达, 等. 应用麻雀搜索和概率神经网络的储能电池故障诊断[J]. 科学技术与工程, 2024, 24(3): 1083-1090.
	YU Siwei, ZHANG Xuesong, LIN Da, et al. Energy storage battery fault diagnosis using Sparrow Search and probabilistic neural network[J]. Sci Tech Engi, 2024, 24(3): 1083-1090. (in Chinese)
[15]	ZHANG Caiping, JIANG Jiuchun, GAO Yang, et al. Charging optimization in lithium-ion batteries based on temperature rise and charge time[J]. Appl Energ, 2017, 194: 569-577.
[16]	GAO Wei, LI Xiaoyu, MA Meina, et al. Case study of an electric vehicle battery thermal runaway and online internal short-circuit detection[J]. IEEE Trans Power Elect, 2021, 36(3): 2452-2455.
[17]	ZHANG Chun, ZHOU Xiaping, JIANG Jiuchun, et al. Micro-fault diagnosis of electric vehicle batteries based on the evolution of battery consistency relative position[J]. J Energ Stor, 2022, 52: Paper No 104746.
[18]	JIANG Jiuchun, LI Taiyu, ZHANG Chun, et al. Fault diagnosis method for lithium-ion batteries in electric vehicles based on isolated forest algorithm[J]. J Energ Stor, 2022, 50: Paper No 104177.
[19]	QIU Yishu, DONG Di, LIN Da, et al. Fault diagnosis for lithium-ion battery energy storage systems based on local outlier factor[J]. J Energ Stor, 2022, 55: 105470.
[20]	王健俊, 陈豪, 付元承. 基于集成学习与数据驱动的电动汽车动力电池多维度故障预警[J]. 汽车安全与节能学报, 2024, 15(3): 368-378.
	WANG Jianjun, CHEN Hao, FU Yuancheng. Multi-dimensional fault warning of electric vehicle power battery based on integrated learning and data drive[J]. J Autom Safe Energ, 2024, 15(3): 368-378. (in Chinese)
[21]	单丰武, 汪超仁, 曾建邦, 等. 基于OPTICS聚类和电压异常指数的动力电池单体不一致性故障程度分析方法[J]. 中国科学: 技术科学, 2024, 54(8): 1611-1624.
	SHAN Fengwu, WANG Chaoren, ZENG Jianbang, et al. Inconsistent fault degree analysis method of power cell based on OPTICS clustering and voltage anomaly index[J]. **Sci China: Tech Sci**, 2024, 54(8): 1611-1624. (in Chinese)
[22]	CHENG Xianfu, LI Xiaojing, MA Xiaodong. A method for battery fault diagnosis and early warning combining isolated forest algorithm and sliding window[J]. Energ Sci Engi, 2023, 11(12): 4493-4504.
[23]	魏正新, 吕晗珺, 闵永军, 等. 基于局部离群点检测的动力电池组不一致早期故障预警[J]. 重庆理工大学学报(自然科学), 2024, 38(6): 21-29.
	WEI Zhengxin, LÜ Hanjun, MIN Yongjun, et al. Inconsistent early fault warning of power battery pack based on local outlier detection[J]. J Chongqing Univ of Tech (Nat Sci), 2024, 38(6): 21-29. (in Chinese)

方法	引文	基础模型	准确率/ %	预警提前时间	时间复杂度
集成学习	[20]	SVM、LightGBM、XGBoost	93.7	-	O(n²)—O(n³)
聚类	[21]	OPTICS聚类	97.3	2~7天	O(n lb n)—O(n²)
无监督学习	[22]	孤立森林	91.0	0.2~3.6 h	O(n)
离群点检测	[23]	局部异常因子	-	10~37天	O(n lb n)
本文方法	-	DBSCAN聚类	96.7	51天	O(n lb n)

方法	引文	基础模型	准确率/ %	预警提前时间	时间复杂度
集成学习	[20]	SVM、LightGBM、XGBoost	93.7	-	O(n²)—O(n³)
聚类	[21]	OPTICS聚类	97.3	2~7天	O(n lb n)—O(n²)
无监督学习	[22]	孤立森林	91.0	0.2~3.6 h	O(n)
离群点检测	[23]	局部异常因子	-	10~37天	O(n lb n)
本文方法	-	DBSCAN聚类	96.7	51天	O(n lb n)