基于COA-LSTM和VMD的锂离子电池剩余寿命预测

doi:10.19799/j.cnki.2095-4239.2024.0157

摘要/Abstract

摘要：

电动汽车中的动力电池在其使用期间寿命的退化是不可避免的，因此研究电动汽车锂电池的使用寿命与利用效率具有重要意义。锂离子电池的剩余使用寿命(remaining useful life, RUL)是表征电池性能的一个重要指标。本文提出了基于郊狼算法优化长短期记忆网络(long short-term memory, LSTM)的动力电池RUL预测模型。首先，对锂离子电池的容量衰减特性进行分析，基于动力电池充放电曲线，提取等恒流充放电间隔、等压升时间间隔作为间接健康因子并通过Pearson法对其进行相关性分析。本文提出变分模态分解(variational mode decomposition, VMD)对健康因子进行分解，得到模态分量。采用LSTM作为动力电池模型来预测RUL，针对LSTM模型参数不精确会影响RUL的预测精度，提出郊狼优化算法(Coyote optimization algorithm, COA)对LSTM模型参数进行优化，以提升模型的预测能力。最后，基于NASA研究中心的公开数据集，将所提方法与LSTM、VMD-LSTM，高斯过程回归(Gaussian process regression, GPR)，BP神经网络模型(backpropagation neural network)进行对比，对COA-LSTM模型的准确性进行验证。实验结果表明，RUL预测误差在2.1%以内，所提方法能够精确预测RUL。

关键词: 锂离子电池, 剩余使用寿命, 郊狼优化算法, 长短期记忆网络

Abstract:

Degradation of battery packs in electric vehicles is inevitable during their operational lifetime, making the estimated remaining useful life (RUL) a critical indicator of battery performance. This study proposes an optimized long short-term memory (LSTM) network-based RUL prediction model for EV lithium-ion batteries using the coyote optimization algorithm (COA). First, this study examines the capacity degradation characteristics of lithium-ion batteries. Indirect health indicators were extracted from the charge and discharge curves of the batteries, including constant current charging and discharging intervals and constant voltage holding time intervals. The correlations of these indicators were examined using the Pearson approach. Then, variational mode decomposition (VMD) was applied to decompose the health indicators into modal components. The LSTM model was used to predict the RUL of the battery pack. To address the issue of inaccurate LSTM model parameters affecting RUL prediction accuracy, COA was used to optimize these parameters and enhance the predictive capabilities of the model. The proposed method was validated using publicly available datasets from the NASA research center and compared with LSTM, VMD-LSTM, Gaussian process regression, and backpropagation neural network models. The experimental results indicate that the proposed approach can achieve RUL prediction errors of within 2%, demonstrating its ability to accurately predict RUL.

Key words: lithium-ion battery, remaining useful life, coyote optimization algorithm, long short-term memory

中图分类号:

TM 912

孙中麟, 李嘉波, 田迪, 王志璇, 邢晓静. 基于COA-LSTM和VMD的锂离子电池剩余寿命预测[J]. 储能科学与技术, 2024, 13(9): 3254-3265.

Zhonglin SUN, Jiabo LI, Di TIAN, Zhixuan WANG, Xiaojing XING. Useful life prediction for lithium-ion batteries based on COA-LSTM and VMD[J]. Energy Storage Science and Technology, 2024, 13(9): 3254-3265.

图/表 29

图1

图2

表1

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

图14

表2

表3

表4

表5

图15

图16

图17

图18

图19

表6

表7

表8

表9

图20

参考文献 14

1	TANG X P, ZOU C F, WIK T, et al. Run-to-Run control for active balancing of lithium iron phosphate battery packs[J]. IEEE Transactions on Power Electronics, 2020, 35(2): 1499-1512. DOI: 10.1109/TPEL.2019.2919709.
2	林娜, 朱武, 邓安全. 基于融合方法预测锂离子电池剩余寿命[J]. 科学技术与工程, 2020, 20(5): 1928-1933. DOI: 10.3969/j.issn.1671-1815.2020.05.033.
	LIN N, ZHU W, DENG A Q. Remaining useful life prediction of the lithium-ion battery based on fusion method[J]. Science Technology and Engineering, 2020, 20(5): 1928-1933. DOI: 10.3969/j.issn.1671-1815.2020.05.033.
3	张宇波, 王有元, 黄洞宁, 等. 面向变工况条件的锂离子电池寿命退化预测方法[J]. 储能科学与技术, 2023, 12(7): 2238-2245. DOI: 10.19799/j.cnki.2095-4239.2023.0233.
	ZHANG Y B, WANG Y Y, HUANG D N, et al. Prognostic method of lithium-ion battery lifetime degradation under various working conditions[J]. Energy Storage Science and Technology, 2023, 12(7): 2238-2245. DOI: 10.19799/j.cnki.2095-4239.2023.0233.
4	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: 109057. DOI: 10.1016/j.measurement.2021.109057.
5	LIU C, WANG Y J, CHEN Z H. Degradation model and cycle life prediction for lithium-ion battery used in hybrid energy storage system[J]. Energy, 2019, 166: 796-806. DOI: 10.1016/j.energy.2018.10.131.
6	刘月峰, 赵光权, 彭喜元. 多核相关向量机优化模型的锂电池剩余寿命预测方法[J]. 电子学报, 2019, 47(6): 1285-1292. DOI: 10.3969/j.issn.0372-2112.2019.06.015.
	LIU Y F, ZHAO G Q, PENG X Y. A lithium-ion battery remaining using life prediction method based on multi-kernel relevance vector machine optimized model[J]. Acta Electronica Sinica, 2019, 47(6): 1285-1292. DOI: 10.3969/j.issn.0372-2112.2019.06.015.
7	OUYANG T C, XU P H, CHEN J X, et al. An online prediction of capacity and remaining useful life of lithium-ion batteries based on simultaneous input and state estimation algorithm[J]. IEEE Transactions on Power Electronics, 2021, 36(7): 8102-8113. DOI: 10.1109/TPEL.2020.3044725.
8	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: 228450. DOI: 10.1016/j.jpowsour.2020.228450.
9	陈翔, 夏飞. 基于CEEMD-AKF的锂电池剩余使用寿命预测方法[J]. 哈尔滨理工大学学报, 2023, 28(3): 28-36. DOI: 10.15938/j.jhust.2023.03.004.
	CHEN X, XIA F. Remaining useful life predictionmethod for lithium-ion batteries based on CEEMD-AKF[J]. Journal of Harbin University of Science and Technology, 2023, 28(3): 28-36. DOI: 10.15938/j.jhust.2023.03.004.
10	REN L, ZHAO L, HONG S, et al. Remaining useful life prediction for lithium-ion battery: A deep learning approach[J]. IEEE Access, 2018, 6: 50587-50598. DOI: 10.1109/ACCESS.2018.2858856.
11	SHU X, LI G, SHEN J W, et al. A uniform estimation framework for state of health of lithium-ion batteries considering feature extraction and parameters optimization[J]. Energy, 2020, 204: 117957. DOI: 10.1016/j.energy.2020.117957.
12	ZHANG Y Z, XIONG R, HE H W, et al. Long short-term memory recurrent neural network for remaining useful life prediction of lithium-ion batteries[J]. IEEE Transactions on Vehicular Technology, 2018, 67(7): 5695-5705. DOI: 10.1109/TVT.2018.2805189.
13	徐彬泰, 孟祥鹿, 田安琪, 等. 基于粒子群优化及高斯过程回归的铅酸电池荷电状态预测[J]. 南京理工大学学报, 2018, 42(2): 162-168. DOI: 10.14177/j.cnki.32-1397n.2018.42.02.005.
	XU B T, MENG X L, TIAN A Q, et al. Prediction for state of charge of lead-acid battery by particle swarm optimization with Gaussian process regression[J]. Journal of Nanjing University of Science and Technology, 2018, 42(2): 162-168. DOI: 10.14177/j.cnki.32-1397n.2018.42.02.005.
14	张浩, 胡昌华, 杜党波, 等. 多状态影响下基于Bi⁃LSTM网络的锂电池剩余寿命预测方法[J]. 电子学报, 2022, 50(3): 619-624. DOI: 10.12263/DZXB.20210207.
	ZHANG H, HU C H, DU D B, et al. Remaining useful life prediction method of lithium? Ion battery based on Bi-LSTM network under Multi-state influence[J]. Acta Electronica Sinica, 2022, 50(3): 619-624. DOI: 10.12263/DZXB.20210207.

Spearman	CCCT	CCDT	CVBT
B05	0.962	0.999	0.959
B06	0.967	0.999	0.965
B07	0.964	0.999	0.958
B18	0.956	0.999	0.954

电池	评价指标	LSTM	VMD-LSTM	VMD-COA-LSTM	GPR	BP
B05	RMSE	0.039581	0.034767	0.01410	0.019638	0.02653
	MAE	0.034842	0.031858	0.01102	0.018524	0.020902
	MAPE	0.025938	0.023772	0.00970	0.013363	0.016052

电池	评价指标	LSTM	VMD-LSTM	VMD-COA-LSTM	GPR	BP
B06	RMSE	0.074957	0.054723	0.01570	0.034601	0.043288
	MAE	0.062265	0.046706	0.01167	0.027603	0.034732
	MAPE	0.048457	0.036475	0.00978	0.020485	0.026712

电池	评价指标	LSTM	VMD-LSTM	VMD-COA-LSTM	GPR	BP
B07	RMSE	0.047274	0.037806	0.00983	0.022525	0.025216
	MAE	0.041957	0.033707	0.01641	0.022055	0.020375
	MAPE	0.028577	0.024875	0.01021	0.014722	0.013874

电池	评价指标	LSTM	VMD-LSTM	VMD-COA-LSTM	GPR	BP
B18	RMSE	0.02626	0.017482	0.01023	0.022467	0.022985
	MAE	0.02106	0.014850	0.01198	0.022462	0.020549
	MAPE	0.01512	0.01074	0.01134	0.015381	0.014672