基于主成分分析与WOA-Elman的锂离子电池SOH估计

doi:10.19799/j.cnki.2095-4239.2022.0384

摘要/Abstract

摘要：

准确估计锂离子电池健康状态(state of health，SOH)是保证电动汽车高效安全持久运行的关键。利用数据驱动方法可以提高SOH估计的精度，然而该方法的SOH估计精度高度依赖于所选择的特征与估计模型。特征之间的冗余性和估计模型泛化性不足都将影响电池SOH的准确估计。为了减小数据驱动特征之间的冗余度，增加模型的泛化性并提升SOH估计精度，提出了一种基于主成分分析与鲸鱼优化算法(whale optimization algorithm，WOA)-Elman的SOH估计方法。首先，从充电曲线中提取并选择与锂离子电池老化高度相关的特征，利用主成分分析方法进行特征降维，减小特征之间的冗余度，然后，采用WOA方法优化Elman模型的初始权值与初始阈值，建立WOA-Elman模型，以B01号电池测试数据训练模型，利用B02与B03号电池进行验证，同时，对比常用的长短期记忆神经网络、支持向量回归和极限学习机以及未优化的Elman模型，结果显示，WOA-Elman估计模型的均方根误差为1.2113%。最后，分别采用3组电池实验测试数据交替作为训练集，对其余两组电池的SOH进行估计验证，估计结果的均方根偏差最大仅为0.1771%。因此，本工作的方法可以更准确地估计电池SOH，并且具有更好的泛化性能。

关键词: 锂离子电池, SOH估计, 主成分分析, Elman神经网络, 鲸鱼优化算法

Abstract:

Accurate estimation state of health (SOH) of lithium-ion batteries is the key to ensuring the efficient, safe, and sustainable operation of electric vehicles. The accuracy of SOH estimation can be improved with a data-driven method. However, the accuracy of SOH estimation in this method is highly dependent on the selected feature and estimation model. The redundancy between features and the lack of generalization of the estimation model will affect the accurate estimation of battery SOH. According to principal component analysis and whale optimization algorithm (WOA)-Elman, a new SOH estimation method is proposed to reduce the redundancy of input features, increase the model's generalization, and improve the accuracy of SOH estimation. Firstly, the features highly related to the aging of lithium-ion batteries were extracted and selected from the charging process curve. Principal component analysis was used to decrease the dimension of features and reduce the redundancy between features. Then, the WOA method was used to optimize the initial weights and thresholds of the Elman model to establish the WOA-Elman model. The B01 battery was used to train the model while B02 and B03 batteries were used to verify the model. Simultaneously, comparing the commonly used long short-term memory neural network, support vector regression, extreme learning machine, and the unoptimized Elman model. The results show that the root-mean-square error of the WOA-Elman estimation model is 1.2113%. Finally, the SOH of the remaining two groups of batteries was estimated and verified by alternating test data of three groups of batteries as training sets, and the maximum root-mean-square deviation of the estimated results was only 0.1771%. Therefore, the proposed method can estimate battery SOH more accurately and perform better generalization.

Key words: lithium-ion battery, state of health estimation, principal component analysis, elman neural network, whale optimization algorithm

中图分类号:

TM 912

李旭东, 张向文. 基于主成分分析与WOA-Elman的锂离子电池SOH估计[J]. 储能科学与技术, 2022, 11(12): 4010-4021.

Xudong LI, Xiangwen ZHANG. State of health estimation method for lithium-ion batteries based on principal component analysis and whale optimization algorithm-Elman model[J]. Energy Storage Science and Technology, 2022, 11(12): 4010-4021.

图/表 24

图1

图2

表1

图3

图4

图5

图6

图7

表2

表3

表4

表5

表6

表7

表8

图8

图9

表9

B02号锂离子电池的误差量化结果"

SOH估计模型	误差量化标准/%
SOH估计模型	$R M S E$	$M A P E$	$M A E$
WOA-Elman	1.2113	1.1591	0.9968
Elman	1.6132	1.5463	1.3182
LSTM	1.6226	1.3329	1.1573
SVR	1.6562	1.5332	1.3007
ELM	1.6477	1.5840	1.3497

表9

表10

B03号锂离子电池的误差量化结果"

SOH估计模型	误差量化标准/%
SOH估计模型	$R M S E$	$M A P E$	$M A E$
WOA-Elman	0.9057	0.6949	0.6032
Elman	0.9921	0.8418	0.7257
LSTM	1.4986	1.0121	0.9024
SVR	1.0157	0.8724	0.7530
ELM	1.0109	0.8674	0.7503

表10

图10

图11

图12

图13

表11

参考文献 26

1	TANG X P, LIU K L, WANG X, et al. Model migration neural network for predicting battery aging trajectories[J]. IEEE Transactions on Transportation Electrification, 2020, 6(2): 363-374.
2	XIONG R, LI L L, TIAN J P. Towards a smarter battery management system: A critical review on battery state of health monitoring methods[J]. Journal of Power Sources, 2018, 405: 18-29.
3	梁新成, 张勉, 黄国钧. 基于BMS的锂离子电池建模方法综述[J]. 储能科学与技术, 2020, 9(6): 1933-1939.
	LIANG X C, ZHANG M, HUANG G J. Review on lithium-ion battery modeling methods based on BMS[J]. Energy Storage Science and Technology, 2020, 9(6): 1933-1939.
4	周頔, 宋显华, 卢文斌, 等. 基于日常片段充电数据的锂电池健康状态实时评估方法研究[J]. 中国电机工程学报, 2019, 39(1): 105-111, 325.
	ZHOU D, SONG X H, LU W B, et al. Real-time SOH estimation algorithm for lithium-ion batteries based on daily segment charging data[J]. Proceedings of the CSEE, 2019, 39(1): 105-111, 325.
5	BIAN X L, WEI Z B, LI W H, et al. State-of-health estimation of lithium-ion batteries by fusing an open circuit voltage model and incremental capacity analysis[J]. IEEE Transactions on Power Electronics, 2022, 37(2): 2226-2236.
6	LI J, ADEWUYI K, LOTFI N, et al. A single particle model with chemical/mechanical degradation physics for lithium ion battery state of health (SOH) estimation[J]. Applied Energy, 2018, 212: 1178-1190.
7	HU X S, CHE Y H, LIN X K, et al. Battery health prediction using fusion-based feature selection and machine learning[J]. IEEE Transactions on Transportation Electrification, 2021, 7(2): 382-398.
8	LI W H, SENGUPTA N, DECHENT P, et al. Online capacity estimation of lithium-ion batteries with deep long short-term memory networks[J]. Journal of Power Sources, 2021, 482: doi: 10.1016/j.jpowsour.2020.228863.
9	WEI J W, DONG G Z, CHEN Z H. Remaining useful life prediction and state of health diagnosis for lithium-ion batteries using particle filter and support vector regression[J]. IEEE Transactions on Industrial Electronics, 2018, 65(7): 5634-5643.
10	PAN H H, LÜ Z Q, WANG H M, et al. Novel battery state-of-health online estimation method using multiple health indicators and an extreme learning machine[J]. Energy, 2018, 160: 466-477.
11	李超然, 肖飞, 樊亚翔, 等. 基于卷积神经网络的锂离子电池SOH估算[J]. 电工技术学报, 2020, 35(19): 4106-4119.
	LI C R, XIAO F, FAN Y X, et al. An approach to lithium-ion battery SOH estimation based on convolutional neural network[J]. Transactions of China Electrotechnical Society, 2020, 35(19): 4106-4119.
12	徐超, 李立伟, 杨玉新, 等. 基于改进粒子滤波的锂电池SOH预测[J]. 储能科学与技术, 2020, 9(6): 1954-1960.
	XU C, LI L W, YANG Y X, et al. Lithium-ion battery SOH estimation based on improved particle filter[J]. Energy Storage Science and Technology, 2020, 9(6): 1954-1960.
13	郭永芳, 黄凯, 李志刚. 基于短时搁置端电压压降的快速锂离子电池健康状态预测[J]. 电工技术学报, 2019, 34(19): 3968-3978.
	GUO Y F, HUANG K, LI Z G. Fast state of health prediction of lithium-ion battery based on terminal voltage drop during rest for short time[J]. Transactions of China Electrotechnical Society, 2019, 34(19): 3968-3978.
14	YANG D, ZHANG X, PAN R, et al. A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curve[J]. Journal of Power Sources, 2018, 384: 387-395.
15	LI Y Y, STROE D I, CHENG Y H, et al. On the feature selection for battery state of health estimation based on charging-discharging profiles[J]. Journal of Energy Storage, 2021, 33: doi: 10.1016/j.est.2020.102122.
16	戴彦文, 于艾清. 基于健康特征参数的CNN-LSTM＆GRU组合锂电池SOH估计[J]. 储能科学与技术, 2022, 11(5): 1641-1649.
	DAI Y W, YU A Q. Combined CNN-LSTM and GRU based health feature parameters for lithium-ion batteries SOH estimation[J]. Energy Storage Science and Technology, 2022, 11(5): 1641-1649.
17	陈翌, 白云飞, 何瑛. 数据驱动的锂电池健康状态估算方法比较[J]. 储能科学与技术, 2019, 8(6): 1204-1210.
	CHEN Y, BAI Y F, HE Y. Comparison of data-driven lithium battery state of health estimation methods[J]. Energy Storage Science and Technology, 2019, 8(6): 1204-1210.
18	李练兵, 李思佳, 李洁, 等. 基于差分电压和Elman神经网络的锂离子电池RUL预测方法[J]. 储能科学与技术, 2021, 10(6): 2373-2384.
	LI L B, LI S J, LI J, et al. RUL prediction of lithium-ion battery based on differential voltage and Elman neural network[J]. Energy Storage Science and Technology, 2021, 10(6): 2373-2384.
19	LI X Y, ZHANG L, WANG Z P, et al. Remaining useful life prediction for lithium-ion batteries based on a hybrid model combining the long short-term memory and Elman neural networks[J]. Journal of Energy Storage, 2019, 21: 510-518.
20	胡兵, 詹仲强, 陈洁, 等. 基于PCA-GA-Elman的短期光伏出力预测研究[J]. 太阳能学报, 2020, 41(6): 256-263.
	HU B, ZHAN Z Q, CHEN J, et al. Prediction research on short-term photovoltaic output based on PCA-GA-Elman[J]. Acta Energiae Solaris Sinica, 2020, 41(6): 256-263.
21	GUO P Y, CHENG Z, YANG L. A data-driven remaining capacity estimation approach for lithium-ion batteries based on charging health feature extraction[J]. Journal of Power Sources, 2019, 412: 442-450.
22	MIRJALILI S, LEWIS A. The whale optimization algorithm[J]. Advances in Engineering Software, 2016, 95: 51-67.
23	潘海鸿, 吕治强, 付兵, 等. 采用极限学习机实现锂离子电池健康状态在线估算[J]. 汽车工程, 2017, 39(12): 1375-1381, 1396.
	PAN H H, LÜ Z Q, FU B, et al. Online estimation of lithium-ion battery's state of health using extreme learning machine[J]. Automotive Engineering, 2017, 39(12): 1375-1381, 1396.
24	石伟杰, 王海民. 基于锂离子电池热特性的SOH在线诊断模型研究[J]. 仪器仪表学报, 2020, 41(8): 206-216.
	SHI W J, WANG H M. On-line diagnosis model of SOH based on thermal characteristics of lithium-ion battery[J]. Chinese Journal of Scientific Instrument, 2020, 41(8): 206-216.
25	徐佳宁, 倪裕隆, 朱春波. 基于改进支持向量回归的锂电池剩余寿命预测[J]. 电工技术学报, 2021, 36(17): 3693-3704.
	XU J N, NI Y L, ZHU C B. Remaining useful life prediction for lithium-ion batteries based on improved support vector regression[J]. Transactions of China Electrotechnical Society, 2021, 36(17): 3693-3704.
26	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.

阶段	特征
CC阶段	T_CC、V_CC、T_DVF
CV阶段	T_CV、I_CV、T_DIF

特征	皮尔森相关系数
特征	B01	B02	B03
T_CC	0.9777	0.9804	0.9832
V_CC	-0.9622	-0.9680	-0.9696
T_DVF	0.9747	0.9778	0.9806
T_CV	-0.9568	-0.9607	-0.9690
I_CV	0.9577	0.9649	0.9663
T_DIF	-0.9600	-0.9670	-0.9686

特征	VIF
特征	B01	B02	B03
T_CC	517	587	740
V_CC	142	184	154
T_DVF	998	1146	1213
T_CV	69	50	57
I_CV	25	23	22
T_DIF	49	41	40

主成分	贡献率/%
主成分	B01	B02	B03
主成分1	97.958	97.890	98.086
主成分2	1.196	1.202	1.056
主成分3	0.468	0.497	0.463
主成分4	0.231	0.257	0.251
主成分5	0.136	0.145	0.134
主成分6	0.001	0.009	0.008

电池编号	皮尔森相关系数
B01	-0.9749
B02	-0.9802
B03	-0.9823