一种基于斯皮尔曼秩相关结合神经网络的电池组内部短路故障检测算法

高明裕; 蔡林辉; 孙长城; 刘才明; 张照娓; 董哲康; 何志伟; 高伟伟

doi:10.11999/JEIT210975

一种基于斯皮尔曼秩相关结合神经网络的电池组内部短路故障检测算法

doi: 10.11999/JEIT210975

高明裕^{1, 2},
蔡林辉^{1, 2},
孙长城^{1, 2},
刘才明^{1, 2},
张照娓^{1, 2},
董哲康^{1, 2, 3, ,},
何志伟^{1, 2},
高伟伟⁴

1.
杭州电子科技大学电子信息学院杭州 310018
2.
装备电子研究重点实验室杭州 310018
3.
浙江大学电气工程学院杭州 310027
4.
天能电池集团股份有限公司长兴 313100

基金项目: 国家重点研发计划 (2020YFB1710600)，国家自然科学基金(62171170)，浙江省重点研发计划 (2021C01111)

详细信息

作者简介:
高明裕：男，教授，研究方向为电池组故障检测

蔡林辉：男，硕士生，研究方向为电池组故障检测

孙长城：男，博士生，研究方向为电池组故障检测

刘才明：男，硕士生，研究方向为电池组故障检测

张照娓：女，博士生，研究方向为汽车电子技术

董哲康：男，副教授，研究方向为神经形态计算系统、电路故障诊断

何志伟：男，教授，研究方向为电池组故障检测

高伟伟：女，中级工程师，研究方向为电池组故障检测

通讯作者:
董哲康　englishp@hdu.edu.cn

中图分类号: TM911
计量
- 文章访问数: 630
- HTML全文浏览量: 377
- PDF下载量: 160
- 被引次数: 0
出版历程
- 收稿日期: 2021-09-14
- 修回日期: 2022-03-22
- 录用日期: 2022-03-10
- 网络出版日期: 2022-03-21
- 刊出日期: 2022-11-14

An Internal Short Circuit Fault Detecting of Battery Pack Based on Spearman Rank Correlation Combined with Neural Network

GAO Mingyu^{1, 2},
CAI Linhui^{1, 2},
SUN Changcheng^{1, 2},
LIU Caiming^{1, 2},
ZHANG Zhaowei^{1, 2},
DONG Zhekang^{1, 2, 3
, ,},
HE Zhiwei^{1, 2},
GAO Weiwei⁴

1.
School of Electronic Information, Hangzhou Dianzi University, Hangzhou 310018, China
2.
Key Laboratory of Equipment Electronics Research, Hangzhou 310018, China
3.
College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China
4.
Tianneng Battery Group Co., Ltd, Changxing 313100, China

Funds: The National Key R & D Program of China (2020YFB1710600), The National Natural Science Foundation of China (62171170), The Key R&D Program of Zhejiang Province (2021C01111)

摘要

摘要: 电池组是电动汽车能源系统的重要组成部分，保障其安全性对电动汽车的智能化发展和人的生命财产都具有重要的意义，检测和保障能源系统中电池组的安全性已成为动力电池领域内的研究热点。神经网络被应用于电池组的各项数据检测中，但在电池组内部短路故障中基于相关系数等信号处理的方法仍广泛使用，其实现方案往往存在针对特定对象、需要特定环境、泛用性能较差等问题。基于此，该文融合相关系数和神经网络的特点，提出一种基于斯皮尔曼秩相关结合三通道卷积双向门控循环神经网络(TBi-GRU)的电池组内部短路故障检测算法。首先，基于斯皮尔曼秩相关系数，滑动窗口联合无量纲化，标准化多维度的电池组运行特征；接着利用提取的正常状态下电池组运行特征训练TBi-GRU神经网络；然后基于已训练好的TBi-GRU模型检测内部短路状态下的电池组运行特征，结合预测结果与各通道的动态阈值对电池组状况进行检测。通过理想条件的仿真分析与实际环境的平台验证，验证了该方法能够充分结合斯皮尔曼秩相关系数的鲁棒性强和TBi-GRU神经网络泛用性强的特点，识别出电池组的内部短路故障。
- 内部短路检测 /
- 电池组 /
- 相关系数 /
- 动态阈值
Abstract: Battery pack is an important part of the energy system of electric vehicles. Ensuring its safety is of great significance to the intelligent development of electric vehicles and human life and property. Detecting and guaranteeing the safety of battery pack in the energy system has become a research hotspot in the field of power batteries. Neural network is widely used in battery data detection, but the signal processing method based on correlation coefficient is still widely used in battery short circuit fault, and its implementation scheme often has some problems, such as targeting specific objects, requiring specific environment, and poor performance in general use. Based on this, this paper combines the characteristics of correlation coefficient and neural network, a neural network fault detection algorithm for internal short circuit in battery packs based on Three-channel parallel Bidirectional Gating Recurrent Unit (TBi-GRU) is proposed. Firstly, based on Spearman's rank correlation coefficient, the sliding window is combined with dimensionless and standardized multi-dimensional battery pack operating characteristics. Then, the TBi-GRU neural network is trained by using the extracted operating characteristics of the battery in the normal state. Then, based on the trained TBi-GRU model, the operating characteristics of the battery packs under the internal short circuit state are detected, and the condition of the battery string is detected by combining the prediction results with the dynamic thresholds of each channel. Through simulation analysis of ideal conditions and platform verification of actual environment, it is proved that this method can fully combine the strong robustness of Szpilman's rank correlation coefficient and the strong universality of TBI-GRU neural network to identify accurately the battery pack's internal short circuit fault.
- Internal short circuit detecing /
- Battery pack /
- Correlation coefficient /
- Dynamic threshold

HTML全文

图 1 整体架构

下载: 全尺寸图片幻灯片

图 2 电池组故障特征提取的过程

下载: 全尺寸图片幻灯片

图 3 模型网络结构

下载: 全尺寸图片幻灯片

图 4 可视化模型输入和输出的计算过程

下载: 全尺寸图片幻灯片

图 5 故障检测过程

下载: 全尺寸图片幻灯片

图 6 10 Ω仿真实验电压响应及局部放大图

下载: 全尺寸图片幻灯片

图 7 5 Ω仿真实验电压响应及局部放大图

下载: 全尺寸图片幻灯片

图 8 1 Ω仿真实验电压响应及局部放大图

下载: 全尺寸图片幻灯片

图 9 模型预测结果图

下载: 全尺寸图片幻灯片

图 10 不同ISC时间的电压响应和特征提取结果对比

下载: 全尺寸图片幻灯片

图 11 锂离子模拟ISC实验设置

下载: 全尺寸图片幻灯片

图 12 10 Ω短路模型前后对比图

下载: 全尺寸图片幻灯片

图 13 5 Ω短路模型前后对比图

下载: 全尺寸图片幻灯片

图 14 1 Ω短路模型前后对比图

下载: 全尺寸图片幻灯片

图 15 0.5 A恒流放电过程对比

下载: 全尺寸图片幻灯片

表 1 故障检测算法对比

	文献[10]	文献[11]	文献 [12, 13]	文献 [6, 16, 17]	本文
电池参数估计	不需要	不需要	需要	不需要	不需要
特征提取	无	有	无	无	有
模型训练	不需要	不需要	不需要	需要	需要
多通道特性	无	无	无	无	有

下载: 导出CSV

表 2 检测指标的描述

指标		预测		情况描述
指标		P	N	情况描述
实际	P	TP	FN	TP(True Positive)：预测结果是故障，实际结果也是故障
	P	TP	FN	FN( False Negative)：预测结果是故障，真实结果是正常
	N	FP	TN	FP( False Positive)：预测结果是正常，实际结果是故障
	N	FP	TN	TN( True Negative)：预测结果是正常，真实结果是正常

下载: 导出CSV

表 3 仿真的锂离子电池Lithium-lon的性能参数

电池参数	数值
额定电压	3.7 V
额定容量	1.35 Ah
电池内阻	27.4 mΩ
响应时间	30 s

下载: 导出CSV

表 4 对比模型参数设置

参数	模型
参数	RNN	LSTM	GRU	TBi-GRU
隐藏神经单元	(6412864)	(6412864)	(6412864)	(64128)
卷积核	0	0	0	5
激活函数	Sigmoid	Sigmoid	Sigmoid	LeakyReLUs, Sigmoid
优化器	Adam	Adam	Adam	Adam
初始学习率	0.01	0.01	0.01	0.01

下载: 导出CSV

表 5 模型检测结果

方法	10 Ω故障定位	5 Ω故障定位	1 Ω故障定位
DTW+TBi-GRU	无	无	1号
ED+TBi-GRU	无	无	1号
Pearson+TBi-GRU	8号	3号	1号
Spearman+RNN	8号	3号	1号
Spearman+LSTM	8号	3号	1号
Spearman+GRU	8号	3号	1号
Spearman+TBi-GRU	8号	3号	1号

下载: 导出CSV

表 6 模型预测结果指标对比(%)

阻值(Ω)	类别	Recall	Precission	F1_Score	Accuracy
1	ED+TBi-GRU	100.00	49.25	66.00	97.73
	DWT+TBi-GRU	100.00	52.24	68.63	97.87
	Person+TBi-GRU	83.13	98.57	90.20	99.00
	Spearman+RNN	44.53	85.07	58.46	94.60
	Spearman+LSTM	52.59	91.04	66.67	95.93
	Spearman+GRU	50.85	89.55	64.86	95.67
	Spearman+TBi-GRU	79.76	100.00	88.74	98.87
5	ED+TBi-GRU	80.95	28.33	41.98	96.87
	DWT+TBi-GRU	82.14	38.33	52.27	97.20
	Person+TBi-GRU	66.67	93.33	77.78	97.87
	Spearman+RNN	32.22	48.33	38.67	93.87
	Spearman+LSTM	41.41	68.33	51.57	94.93
	Spearman+GRU	40.82	66.67	50.63	94.80
	Spearman+TBi-GRU	84.21	80.00	82.05	98.60
10	ED+TBi-GRU	无	0.00	无	95.53
	DWT+TBi-GRU	无	0.00	无	95.53
	Person+TBi-GRU	72.73	71.64	72.18	97.53
	Spearman+RNN	38.24	38.81	38.52	94.47
	Spearman+LSTM	79.17	28.36	41.76	96.47
	Spearman+GRU	75.00	26.87	39.56	96.33
	Spearman+TBi-GRU	89.66	81.25	85.25	98.80

下载: 导出CSV

表 7 各种阈值方法对比结果(%)

阻值(Ω)	阈值类别	Recall	Precission	F1_Score	Accuracy
1	max	40.12	100.00	57.26	93.33
	μ+3σ	29.17	93.33	44.44	90.67
	本文	79.76	100.00	88.74	98.87
5	max	29.17	93.33	44.44	90.67
	μ+3σ	89.66	43.33	58.43	97.53
	本文	84.21	80.00	82.05	98.60
10	max	52.21	88.06	65.56	95.87
	μ+3σ	89.19	49.25	63.46	97.47
	本文	89.66	81.25	85.25	98.80

下载: 导出CSV

表 8 预测结果与原始数据对噪声和较低ISC的抑制效果(%)对比

方法	抑制较低ISC			抑制噪声
方法	1 Ω	5 Ω	10 Ω	1 Ω	5 Ω	10 Ω
Person+TBi-GRU	–12.53	39.00	19.70	14.94	27.26	15.32
Spearman+RNN	0.62	25.73	9.20	–20.18	–25.44	–19.37
Spearman+LSTM	–0.15	32.40	16.01	–15.10	–19.43	–9.03
Spearman+GRU	–0.74	16.70	18.23	–13.14	–17.08	–7.34
Spearman+TBi-GRU	–4.44	18.65	8.09	12.52	22.61	16.16

下载: 导出CSV

参考文献(26)

[1]	DIAO Weiping, NAQVI I H, and PECHT M. Early detection of anomalous degradation behavior in lithium-ion batteries[J]. Journal of Energy Storage, 2020, 32: 101710. doi: 10.1016/j.est.2020.101710
[2]	SEO M, GOH T, PARK M, et al. Detection method for soft internal short circuit in lithium-ion battery pack by extracting open circuit voltage of faulted cell[J]. Energies, 2018, 11(7): 1669. doi: 10.3390/en11071669
[3]	张照娓, 郭天滋, 高明裕, 等. 电动汽车锂离子电池荷电状态估算方法研究综述[J]. 电子与信息学报, 2021, 43(7): 1803–1815. doi: 10.11999/JEIT200487 ZHANG Zhaowei, GUO Tianzi, GAO Mingyu, et al. Review of SoC estimation methods for electric vehicle li-ion batteries[J]. Journal of Electronics &Information Technology, 2021, 43(7): 1803–1815. doi: 10.11999/JEIT200487
[4]	FENG Xuning, PAN Yue, HE Xiangming, et al. Detecting the internal short circuit in large-format lithium-ion battery using model-based fault-diagnosis algorithm[J]. Journal of Energy Storage, 2018, 18: 26–39. doi: 10.1016/j.est.2018.04.020
[5]	NAHA A, KHANDELWAL A, HARIHARAN K S, et al. On-board short-circuit detection of li-ion batteries undergoing fixed charging profile as in smartphone applications[J]. IEEE Transactions on Industrial Electronics, 2019, 66(11): 8782–8791. doi: 10.1109/TIE.2018.2889623
[6]	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]. Journal of Cleaner Production, 2018, 187: 950–959. doi: 10.1016/j.jclepro.2018.03.259
[7]	KANG Yongzhe, DUAN Bin, ZHOU Zhongkai, et al. A multi-fault diagnostic method based on an interleaved voltage measurement topology for series connected battery packs[J]. Journal of Power Sources, 2019, 417: 132–144. doi: 10.1016/j.jpowsour.2019.01.058
[8]	ZHANG Mingxuan, DU Jiuyu, LIU Lishuo, et al. Internal short circuit detection method for battery pack based on circuit topology[J]. Science China Technological Sciences, 2018, 61(10): 1502–1511. doi: 10.1007/s11431-017-9299-3
[9]	ZHENG Yuejiu, LU Yifan, GAO Wenkai, et al. Micro-short-circuit cell fault identification method for lithium-ion battery packs based on mutual information[J]. IEEE Transactions on Industrial Electronics, 2021, 68(5): 4373–4381. doi: 10.1109/TIE.2020.2984441
[10]	XIA Bing, MI C, CHEN Zheng, et al. Multiple cell lithium-ion battery system electric fault online diagnostics[C]. Proceedings of 2015 IEEE Transportation Electrification Conference and Expo, Dearborn, USA, 2015: 1–7.
[11]	SUN Zhenyu, WANG Zhenpo, LIU Peng, et al. Relative entropy based lithium-ion battery pack short circuit detection for electric vehicle[C]. 2020 IEEE Energy Conversion Congress and Exposition (ECCE), Detroit, USA, 2020: 5061–5067.
[12]	CHIU K C, LIN C H, YEH S F, et al. An electrochemical modeling of lithium-ion battery nail penetration[J]. Journal of Power Sources, 2014, 251: 254–263. doi: 10.1016/j.jpowsour.2013.11.069
[13]	XIONG Rui, YANG Ruixin, CHEN Zeyu, et al. Online fault diagnosis of external short circuit for lithium-ion battery pack[J]. IEEE Transactions on Industrial Electronics, 2020, 67(2): 1081–1091. doi: 10.1109/TIE.2019.2899565
[14]	党翰斌, 马睿, 周俊杰, 等. 基于CNN的质子交换膜燃料电池故障诊断[J]. 电力电子技术, 2020, 54(12): 52–55. doi: 10.3969/j.issn.1000-100X.2020.12.014 DANG Hanbin, MA Rui, ZHOU Junjie, et al. Fault diagnosis of proton exchange membrane fuel cells based on CNN[J]. Power Electronics, 2020, 54(12): 52–55. doi: 10.3969/j.issn.1000-100X.2020.12.014
[15]	彭运赛, 夏飞, 袁博, 等. 基于改进CNN和信息融合的动力电池组故障诊断方法[J]. 汽车工程, 2020, 42(11): 1529–1536. doi: 10.19562/j.chinasae.qcgc.2020.11.011 PENG Yunsai, XIA Fei, YUAN Bo, et al. Fault diagnosis of traction battery pack based on improved convolution neural network and information fusion[J]. Automotive Engineering, 2020, 42(11): 1529–1536. doi: 10.19562/j.chinasae.qcgc.2020.11.011
[16]	NAHA A, KHANDELWAL A, AGARWAL S, et al. Internal short circuit detection in Li-ion batteries using supervised machine learning[J]. Scientific Reports, 2020, 10(1): 1301. doi: 10.1038/s41598-020-58021-7
[17]	YANG Ruixin, XIONG Rui, MA Suxiao, et al. Characterization of external short circuit faults in electric vehicle Li-ion battery packs and prediction using artificial neural networks[J]. Applied Energy, 2020, 260: 114253. doi: 10.1016/j.apenergy.2019.114253
[18]	ZHEN C, CHEN Ziqiang, and HUANZ D. A novel sensor fault diagnosis method for lithium-ion battery system using hybrid system modeling[C]. 2018 Condition Monitoring and Diagnosis (CMD), Perth, Australia, 2018: 1–5.
[19]	XIONG Rui, SUN Wanzhou, YU Quanqing, et al. Research progress, challenges and prospects of fault diagnosis on battery system of electric vehicles[J]. Applied Energy, 2020, 279: 115855. doi: 10.1016/j.apenergy.2020.115855
[20]	WILLIARD N, HE Wei, HENDRICKS C, et al. Lessons learned from the 787 Dreamliner issue on lithium-ion battery reliability[J]. Energies, 2013, 6(9): 4682–4695. doi: 10.3390/en6094682
[21]	XIA Bing, SHANG Yunlong, NGUYEN T, et al. A correlation based fault detection method for short circuits in battery packs[J]. Journal of Power Sources, 2017, 337: 1–10. doi: 10.1016/j.jpowsour.2016.11.007
[22]	STEPHANOU M and VARUGHESE M. Sequential estimation of Spearman rank correlation using Hermite series estimators[J]. Journal of Multivariate Analysis, 2021, 186: 104783. doi: 10.1016/j.jmva.2021.104783
[23]	CHUNG J, GULCEHRE C, CHO K, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[C]. NIPS 2014 Workshop on Deep Learning, Montreal, Canada, 2014. doi: 10.48550/arXiv.1412.3555.
[24]	SU Ya, ZHAO Youjian, NIU Chenhao, et al. Robust anomaly detection for multivariate time series through stochastic recurrent neural network[C]. The 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Anchorage, USA, 2019: 2828–2837.
[25]	徐文畅, 何文明, 游斌权, 等. 基于形态特征提取的急性下壁心肌梗死BiLSTM网络辅助诊断算法[J]. 电子与信息学报, 2021, 43(9): 2561–2568. doi: 10.11999/JEIT200480 XU Wenchang, HE Wenming, YOU Binquan, et al. Acute inferior myocardial infarction detection algorithm based on BiLSTM network of morphological feature extraction[J]. Journal of Electronics &Information Technology, 2021, 43(9): 2561–2568. doi: 10.11999/JEIT200480
[26]	HUNDMAN K, CONSTANTINOU V, LAPORTE C, et al. Detecting spacecraft anomalies using LSTMs and nonparametric dynamic thresholding[C]. The 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, London, UK, 2018: 387–395.