基于数据驱动的动力电池健康状态评估平台

doi:10.19799/j.cnki.2095-4239.2022.0186

摘要/Abstract

摘要：

随着新能源汽车迅速发展普及，对动力电池健康状态进行准确评估成为了一个亟待解决的问题。针对这一问题，本文以轻量梯度提升机(light gradient boosting machine，LightGBM)为基础，采集数据处理特征并进行训练，最终构建了一个动力电池健康状态评估平台。首先，依据相关工作，在原始数据中提取了6个特征工程——放电电压的最小值的时间、负载电压的75百分位、放电电压平均值、放电负载电压的25百分位、放电电压的25百分位和放电电压的标准差。其次通过引入直方图算法、单边梯度采样算法和斥特征捆绑算法，对数据进行进一步处理，以减少平台的内存消耗与计算代价。最后采用NASA艾姆斯卓越预测中心的数据，对平台功能进行验证并与相似工作进行对比，结果表明该平台可以提供良好的预测精度(均方根误差仅0.0103)。该平台对动力电池SOH预测方法的发展具有积极的影响，具有巨大的实际运用潜力。

关键词: 动力电池, 电池健康状态, 评价平台, LightGBM模型

Abstract:

With the rapid development and popularization of electric vehicles (EVs), accurate evaluation of the state of health (SOH) of vehicle power batteries has become a pressing issue. To address the problem, this study adopts a machine learning approach based on the Light Gradient Boosting Machine (LightGBM) framework. This approach involves collecting data, processing characteristics, carrying out training, and finally constructing a data-driven battery SOH analytic system. As a first step, six characteristics are extracted from the original data: the time of minimum discharge voltage; the 75th percentile of load voltage; the average discharge voltage; the 25th percentile of discharge load voltage; the 25th percentile of discharge voltage; and the standard deviation of discharge voltage. Secondly, the data are further processed to reduce memory consumption and computing cost. This processing utilizes LightGBM's key algorithms: histogram-based decision tree learning, Gradient-Based One-Side Sampling, and Exclusive Feature Bundling. Finally, system function is verified and compared with similar work, using data from the NASA Ames Prognostics Center of Excellence. Results demonstrate that the SOH platform can deliver high predictive accuracy (a root-mean-square error of 0.0103). The system represents significant progress in vehicle battery SOH prediction methodology and has high potential for practical application.

Key words: power battery, SOH, evaluation platform, LightGBM

中图分类号:

常泽宇, 张之琦, 张晓东, 李丽, 郁亚娟. 基于数据驱动的动力电池健康状态评估平台[J]. 储能科学与技术, 2022, 11(6): 1847-1853.

CHANG Zeyu, ZHANG Zhiqi, ZHANG Xiaodong, LI Li, YU Yajuan. A data-driven state of health （SOH） assessment platform for vehicle power batteries[J]. Energy Storage Science and Technology, 2022, 11(6): 1847-1853.

图/表 7

图1

图2

图3

图4

图5

图6

表1

参考文献 21

1	左培文, 朱培培, 邵丽青. 新能源汽车动力电池产业发展特点与趋势分析[J]. 汽车文摘, 2022(1): 1-7.
	ZUO P W, ZHU P P, SHAO L Q. The development characteristics and trend analysis of power battery industry for new energy vehicles[J]. Automotive Digest, 2022(1): 1-7.
2	李方方, 张晓龙, 吴怡, 等. 我国动力锂电池行业现状和发展趋势[J]. 交通节能与环保, 2016, 12(3): 14-16.
	LI F F, ZHANG X L, WU Y, et al. Current situation and development trend of power lithium battery industry[J]. Energy Conservation & Environmental Protection in Transportation, 2016, 12(3): 14-16.
3	张颖. 新能源车渗透率继续提高[J]. 汽车与配件, 2021(14): 4.ZHANG Y. New energy vehicle penetration continues to increase[J]. Automobile & Parts, 2021(14): 4.
4	FENG X N, PAN Y, HE X M, 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.
5	熊瑞. 动力电池管理系统核心算法[M]. 北京: 机械工业出版社, 2018: 100-104.
	XIONG R. Core algorithm of battery management system for EVs[M]. Beijing: China Machine Press, 2018: 100-104.
6	GE M F, LIU Y B, JIANG X X, et al. A review on state of health estimations and remaining useful life prognostics of lithium-ion batteries[J]. Measurement, 2021, 174: doi:10.1016/j.measurement.2021.109057.
7	耿萌萌, 范茂松, 杨凯, 等. 基于EIS和神经网络的退役电池SOH快速估计[J]. 储能科学与技术, 2022, 11(2): 673-678.
	GENG M M, FAN M S, YANG K, et al. Fast estimation method for state-of-health of retired batteries based on electrochemical impedance spectroscopy and neural network[J]. Energy Storage Science and Technology, 2022, 11(2): 673-678.
8	韩云飞, 谢佳, 蔡涛, 等. 结合高斯过程回归与特征选择的锂离子电池容量估计方法[J]. 储能科学与技术, 2021, 10(4): 1432-1438.
	HAN Y F, XIE J, CAI T, et al. Capacity estimation of lithium-ion batteries based on Gaussian process regression and feature selection[J]. Energy Storage Science and Technology, 2021, 10(4): 1432-1438.
9	任璞, 王顺利, 何明芳, 等. 基于内阻增加和容量衰减双重标定的锂电池健康状态评估[J]. 储能科学与技术, 2021, 10(2): 738-743.
	REN P, WANG S L, HE M F, et al. State of health estimation of Li-ion battery based on dual calibration of internal resistance increasing and capacity fading[J]. Energy Storage Science and Technology, 2021, 10(2): 738-743.
10	董明, 范文杰, 刘王泽宇, 等. 基于特征频率阻抗的锂离子电池健康状态评估[J/OL]. 中国电机工程学报: 1-11. [2022-04-01]. doi:10.13334/j.0258-8013.pcsee.212036.
	DONG M, FAN W J, LIU Wangzeyu, et al. Health assessment of lithium-ion batteries based on characteristic frequency impedance[J/OL]. Proceedings of the CSEE: 1-11. [2022-04-01]. doi:10.13334/j.0258-8013.pcsee.212036.
11	WANG S L, FERNANDEZ C, YU C M, et al. A novel charged state prediction method of the lithium ion battery packs based on the composite equivalent modeling and improved splice Kalman filtering algorithm[J]. Journal of Power Sources, 2020, 471: doi:10.1016/j.jpowsour.2020.228450.
12	王凡, 史永胜, 刘博亲, 等. 基于注意力改进BiGRU的锂离子电池健康状态估计[J]. 储能科学与技术, 2021, 10(6): 2326-2333.
	WANG F, SHI Y S, LIU B Q, et al. Health state estimation of lithium-ion batteries based on attention augmented BiGRU[J]. Energy Storage Science and Technology, 2021, 10(6): 2326-2333.
13	张孝远, 张金浩, 蒋亚俊. 基于改进TCN模型的动力电池健康状态评估[J]. 储能科学与技术, 2022, 11(5): 1617-1626.
	ZHANG X Y, ZHANG J H, JIANG Y J. Power battery health evaluation based on improved TCN model[J]. Energy Storage Science and Technology, 2022, 11(5): 1617-1626.
14	KE G L, MENG Q, FINLEY T, et al. Lightgbm: A highly efficient gradient boosting decision tree[J]. Advances in Neural Information Processing systems, 2017, 30.
15	MENG Q, KE G L, WANG T F, et al. A communication-efficient parallel algorithm for decision tree[C]//30th Conference on Neural Information Processing Systems (NIPS 2016), 2016.
16	ZHANG Z Q, LI L, LI X, et al. State-of-health estimation for the lithium-ion battery based on gradient boosting decision tree with autonomous selection of excellent features[J]. International Journal of Energy Research, 2022, 46(2): 1756-1765.
17	LIU D T, PANG J Y, ZHOU J B, et al. Data-driven prognostics for lithium-ion battery based on Gaussian Process Regression[C]// Proceedings of the IEEE 2012 Prognostics and System Health Management Conference, 2012.
18	PENNA J A M, NASCIMENTO C L, RODRIGUES L R. Health monitoring and remaining useful life estimation of lithium-ion aeronautical batteries[C]//2012 IEEE Aerospace Conference, 2012.
19	梅雪峰, 赵礼峰. 基于LightGBM的改进表面肌电信号手势识别研究[J]. 计算机与数字工程, 2022, 50(1): 95-99.
	MEI X F, ZHAO L F. Research on improved gesture recognition of surface EMG based on LightGBM[J]. Computer & Digital Engineering, 2022, 50(1): 95-99.
20	刘兴涛, 刘晓剑, 武骥, 等. 基于曲线压缩与XGBoost算法的锂离子电池SOH估计[J/OL]. 吉林大学学报(工学版): 1-8. [2022-04-13]. doi:10.13229/j.cnki.jdxbgxb20210020.
	LIU X T, LIU X J, WU J, et al. Curve compression and XGBoost based state of health estimation method for lithium-ion battery[J/OL]. Journal of Jilin University (Engineering and Technology Edition): 1-8. [2022-04-13]. doi: 10.13229/j.cnki.jdxbgxb20210020.
21	刘运鑫, 姚良忠, 周金辉, 等. 基于LSTM的锂电池储能装置SOC与SOH联合预测[J]. 全球能源互联网, 2022, 5(1): 37-45.
	LIU Y X, YAO L Z, ZHOU J H, et al. Joint prediction of state of charge and state of health based on LSTM for lithium-ion batteries[J]. Journal of Global Energy Interconnection, 2022, 5(1): 37-45.

指标	LightGBM	GBDT	XGBoost	LSTM
RMSE	0.0103	0.0139	0.011	0.0153
R²	99.5	98.98	98.6	—

[1]	周伟, 符冬菊, 刘伟峰, 陈建军, 胡照, 曾燮榕. 废旧磷酸铁锂动力电池回收利用研究进展[J]. 储能科学与技术, 2022, 11(6): 1854-1864.
[2]	王苏杭, 李建林, 李雅欣, 熊俊杰, 曾伟. 锂离子电池系统低温充电策略[J]. 储能科学与技术, 2022, 11(5): 1537-1542.
[3]	张孝远, 张金浩, 蒋亚俊. 基于改进TCN模型的动力电池健康状态评估[J]. 储能科学与技术, 2022, 11(5): 1617-1626.
[4]	王芳, 王峥, 林春景, 张国振, 张贵平, 马天翼, 刘磊, 刘仕强. 新能源汽车动力电池安全失效潜在原因分析[J]. 储能科学与技术, 2022, 11(5): 1411-1418.
[5]	刘春辉, 任宏斌. 基于SOC的动力电池组主动均衡研究[J]. 储能科学与技术, 2022, 11(2): 667-672.
[6]	马天翼, 王伟哲, 林春景, 白广利, 韦振, 刘仕强. 磷酸铁锂电池多阶段利用可行性[J]. 储能科学与技术, 2022, 11(1): 119-126.
[7]	胡建, 林春景, 郝维健, 郑天雷. 动力电池标准体系建设现状及建议[J]. 储能科学与技术, 2022, 11(1): 313-320.
[8]	郑永强, 吴越, 张盼盼, 雷博, 郑耀东. 基于多分支拓扑的梯次利用储能系统电池同期退役协同控制策略[J]. 储能科学与技术, 2021, 10(6): 2283-2292.
[9]	刘霏霏, 鲍荣清, 程贤福, 李骏, 秦武, 杨超峰. 服役工况下车用锂离子动力电池散热方法综述[J]. 储能科学与技术, 2021, 10(6): 2269-2282.
[10]	李鹏, 李立伟, 杨玉新. 基于IBOA-PF的锂电池健康状态预测[J]. 储能科学与技术, 2021, 10(2): 705-713.
[11]	张志鸿, 牟俊彦, 孟玉发. 风冷圆柱形锂离子电池系统热失控扩展特性[J]. 储能科学与技术, 2021, 10(2): 658-663.
[12]	樊彬, 姜成龙, 林春景, 李玉鹏, 余八一, 张晋杰, 张良, 高孟洋, 王伟, 解坤, 常宏. 电动汽车用动力电池系统循环寿命试验[J]. 储能科学与技术, 2021, 10(2): 671-678.
[13]	王海民, 王寓非, 胡峰. 石墨-石蜡复合相变材料的圆柱型动力电池组热管理性能[J]. 储能科学与技术, 2021, 10(1): 210-217.
[14]	沈剑, 黄碧雄, 谢兆康, 李嘉寅, 刘宁宁, 贾志强. FSEC赛车动力电池箱内部结构布置与优化设计[J]. 储能科学与技术, 2020, 9(S1): 31-38.
[15]	何瑛, 钟根鹏, 陈翌. 基于特征组合堆叠融合集成学习的锂离子动力电池SOC估算[J]. 储能科学与技术, 2020, 9(5): 1548-1557.