基于VMD-PSO-GRU模型的锂离子电池剩余寿命预测

doi:10.19799/j.cnki.2095-4239.2022.0491

摘要/Abstract

摘要：

准确预测锂电池的剩余使用寿命(remaining useful life， RUL)对降低电池使用风险和保证系统的安全运行起着非常重要的作用。为了消除电池容量序列受容量再生等影响，提高预测结果的准确性和稳定性，提出了一种基于变分模态分解(variational modal decomposition，VMD)与参数优化的门控循环神经网络(gate recurrent unit，GRU)相结合的RUL预测模型。首先采用VMD算法将锂电池的容量序列分解为一系列平稳分量；然后采用多层GRU网络对各分量进行预测，针对预测结果不稳定的问题，在模型训练前利用粒子群算法(particle swarm optimization， PSO)对GRU模型的参数进行优化；最后叠加各分量的预测值作为最终预测结果。在NASA数据集上对本模型进行了验证，当采用20个已知电池序列数据预测时，预测结果的最大平均绝对百分比误差和均方根误差控制在0.88%和0.0148以内，RUL预测的最大误差不超过2个充电周期，具有较高的鲁棒性和预测精度。

关键词: 锂电池, 剩余寿命预测, 变分模态分解, 粒子群优化算法, 门控循环神经网络

Abstract:

Accurately predicting the remaining useful life (RUL) of lithium batteries plays a significant role in reducing battery risk and ensuring the safe operation of the system. To reduce the influence of battery capacity regeneration, and improve the accuracy and stability of RUL prediction, an integrated prediction model of variational modal decomposition and gate recurrent unit (GRU) network with particle swarm optimization algorithm was proposed. First, the capacity sequence of the lithium battery was decomposed into a series of stationary components using the variational modal decomposition algorithm. Then, the multiple GRU network was used to predict the capacity component on each sub-sequence. To deal with inconsistent prediction results, particle swarm optimization was used to optimize the parameters of the GRU model before model training. Finally, the predicted results of each sub-sequences were integrated as the final battery capacity estimation, followed by the prediction of RUL. The experimental results on the National Aeronautics and Space Administration dataset showed that the maximum mean absolute percentage error and root mean square error of the results were controlled within 0.88% and 0.0148, respectively, and the maximum error of RUL prediction was less than two charging cycles. It has high robustness and prediction accuracy.

Key words: lithium battery, remaining life prediction, variational modal decomposition, particle swarm optimization, gated recurrent neural network

中图分类号:

TM 912

刘芊彤, 邢远秀. 基于VMD-PSO-GRU模型的锂离子电池剩余寿命预测[J]. 储能科学与技术, 2023, 12(1): 236-246.

Qiantong LIU, Yuanxiu XING. Remaining life prediction of lithium-ion battery based on VMD-PSO-GRU model[J]. Energy Storage Science and Technology, 2023, 12(1): 236-246.

图/表 13

图1

图2

图3

图4

表1

GRU神经网络不同层数预测结果"

$m$	$n$	MAE	RMSE	RA
1	15	0.0114	0.0145	0.9911
	20	0.0103	0.0123	0.9919
	25	0.0101	0.0119	0.9923
2	15	0.0116	0.0181	0.9915
	20	0.0087	0.0112	0.9933
	25	0.0085	0.0113	0.9934
3	15	0.0198	0.0216	0.9854
	20	0.0168	0.0211	0.9866
	25	0.0160	0.0208	0.9876

表1

图5

图6

表2

图7

表3

图8

表4

表5

参考文献 27

1	WANG S L, JIN S Y, BAI D K, et al. A critical review of improved deep learning methods for the remaining useful life prediction of lithium-ion batteries[J]. Energy Reports, 2021, 7: 5562-5574.
2	李练兵, 季亮, 祝亚尊, 等. 等效循环电池组剩余使用寿命预测[J]. 工程科学学报, 2020, 42(6): 796-802.
	LI L B, JI L, ZHU Y Z, et al. Investigation of RUL prediction of lithium-ion battery equivalent cycle battery pack[J]. Chinese Journal of Engineering, 2020, 42(6): 796-802.
3	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.
4	KHODADADI SADABADI K, JIN X, RIZZONI G. Prediction of remaining useful life for a composite electrode lithium ion battery cell using an electrochemical model to estimate the state of health[J]. Journal of Power Sources, 2021, 481: doi: 10.1016/j.jpowsour.2020.228861.
5	宋胜, 李云伍, 赵颖, 等. 锂离子电池片段数据的荷电状态估计研究[J]. 电源技术, 2022, 46(7): 734-738.
	SONG S, LI Y W, ZHAO Y, et al. Research on SOC estimation based on fragment data of lithium-ion battery[J]. Chinese Journal of Power Sources, 2022, 46(7): 734-738.
6	谢滟馨, 王顺利, 史卫豪, 等. 一种用于高保真锂电池SOC估计的无迹粒子滤波新方法[J]. 储能科学与技术, 2021, 10(2): 722-731.
	XIE Y X, WANG S L, SHI W H, et al. A new method of unscented particle filter for high-fidelity lithium-ion battery SOC estimation[J]. Energy Storage Science and Technology, 2021, 10(2): 722-731.
7	焦自权, 范兴明, 张鑫, 等. 基于改进粒子滤波算法的锂离子电池状态跟踪与剩余使用寿命预测方法[J]. 电工技术学报, 2020, 35(18): 3979-3993.
	JIAO Z Q, FAN X M, ZHANG X, et al. State tracking and remaining useful life predictive method of Li-ion battery based on improved particle filter algorithm[J]. Transactions of China Electrotechnical Society, 2020, 35(18): 3979-3993.
8	LIU Z Y, HE H J, XIE J, et al. Self-discharge prediction method for lithium-ion batteries based on improved support vector machine[J]. Journal of Energy Storage, 2022, 55: doi:10.1016/j.est.2022.105571.
9	DENG Z W, HU X S, LIN X K, et al. Data-driven state of charge estimation for lithium-ion battery packs based on Gaussian process regression[J]. Energy, 2020, 205: doi:10.1016/j.energy. 2020.118000.
10	何星, 丁有军, 宋丽君, 等. 基于加速鱼群算法的锂离子电池剩余寿命预测[J]. 兵器装备工程学报, 2022, 43(2): 163-169.
	HE X, DING Y J, SONG L J, et al. Residual life prediction for lithium-ion battery based on accelerating AFSA[J]. Journal of Ordnance Equipment Engineering, 2022, 43(2): 163-169.
11	徐帅, 刘雨辰, 周飞. 基于RNN的锂离子电池SOC估算研究进展[J]. 电源技术, 2021, 45(2): 263-269.
	XU S, LIU Y C, ZHOU F. Research progress of SOC estimation of lithium ion batteries based on RNN[J]. Chinese Journal of Power Sources, 2021, 45(2): 263-269.
12	陈诚, 皮志勇, 赵英龙, 等. 基于自适应灾变遗传-循环神经网络的锂离子电池SOC估计[J]. 电气工程学报, 2022, 17(1): 86-94.
	CHEN C, PI Z Y, ZHAO Y L, et al. State of charge estimation with adaptive cataclysm genetic algorithm-recurrent neural network for Li-ion batteries[J]. Journal of Electrical Engineering, 2022, 17(1): 86-94.
13	刘伟霞, 田勋, 肖家勇, 等. 基于混合模型及LSTM的锂电池SOH与剩余寿命预测[J]. 储能科学与技术, 2021, 10(2): 689-694.
	LIU W X, TIAN X, XIAO J Y, et al. Estimation of SOH and remaining life of lithium batteries based on a combination model and long short-term memory[J]. Energy Storage Science and Technology, 2021, 10(2): 689-694.
14	毕贵红, 谢旭, 蔡子龙, 等. 动态条件下基于深度学习的锂电池容量估计[J]. 汽车工程, 2022, 44(6): 868-877, 885.
	BI G H, XIE X, CAI Z L, et al. Capacity estimation of lithium-ion battery based on deep learning under dynamic conditions[J]. Automotive Engineering, 2022, 44(6): 868-877, 885.
15	范元亮, 方略斌, 吴涵, 等. 基于片段充电时间和GRU的锂离子电池健康状态预测方法: CN113917336A[P]. 2022-01-11.
	FAN Y L, FANG L B, WU H, et al. Lithium ion battery health state prediction method based on fragment charging time and GRU: CN113917336A[P]. 2022-01-11.
16	胡天中, 余建波. 基于多尺度分解和深度学习的锂电池寿命预测[J]. 浙江大学学报(工学版), 2019, 53(10): 1852-1864.
	HU T Z, YU J B. Life prediction of lithium-ion batteries based on multiscale decomposition and deep learning[J]. Journal of Zhejiang University (Engineering Science), 2019, 53(10): 1852-1864.
17	CHEN D Q, HONG W C, ZHOU X Z. Transformer network for remaining useful life prediction of lithium-ion batteries[J]. IEEE Access, 10: 19621-19628.
18	魏孟, 王桥, 叶敏, 等. 基于NARX动态神经网络的锂离子电池剩余寿命间接预测[J]. 工程科学学报, 2022, 44(3): 380-388.
	WEI M, WANG Q, YE M, et al. An indirect remaining useful life prediction of lithium-ion batteries based on a NARX dynamic neural network[J]. Chinese Journal of Engineering, 2022, 44(3): 380-388.
19	DING P, LIU X J, LI H Q, et al. Useful life prediction based on wavelet packet decomposition and two-dimensional convolutional neural network for lithium-ion batteries[J]. Renewable and Sustainable Energy Reviews, 2021, 148: doi: 10.1016/j.rser.2021.111287.
20	郑雪莹, 邓晓刚, 曹玉苹. 基于能量加权高斯过程回归的锂离子电池健康状态预测[J]. 电子测量与仪器学报, 2020, 34(6): 63-69.
	ZHENG X Y, DENG X G, CAO Y P. State of health prediction of lithium-ion batteries based on energy-weighted Gaussian process regression[J]. Journal of Electronic Measurement and Instrumentation, 2020, 34(6): 63-69.
21	黄凯, 丁恒, 郭永芳, 等. 基于数据预处理和长短期记忆神经网络的锂离子电池寿命预测[J]. 电工技术学报, 2022, 37(15): 3753-3766.
	HUANG K, DING H, GUO Y F, et al. Prediction of remaining useful life of lithium-ion battery based on adaptive data preprocessing and long short-term memory network[J]. Transactions of China Electrotechnical Society, 2022, 37(15): 3753-3766.
22	陈红霞, 丁国荣, 陈贵词, 等. 联合变分模态分解和长短时记忆网络的锂离子电池健康状态估计[J/OL]. 电源学报:1-13[2022-11-08].http://kns.cnki.net/kcms/detail/12.1420.TM.20220524.1841.008.html.
	CHEN H X, DING G R, CHEN G C, et al. Health state estimation of lithium-ion batteries based on joint variational modal decomposition and long-term and short-term memory networks[J/OL]. Journal of Power Supply: 1-13 [2022-11-08].http://kns.cnki.net/kcms/detail/12.1420.TM.20220524.1841.008.html.
23	王冉, 后麒麟, 石如玉, 等. 基于变分模态分解与集成深度模型的锂电池剩余寿命预测方法[J]. 仪器仪表学报, 2021, 42(4): 111-120.
	WANG R, HOU Q L, SHI R Y, et al. Remaining useful life prediction method of lithium battery based on variational mode decomposition and integrated deep model[J]. Chinese Journal of Scientific Instrument, 2021, 42(4): 111-120.
24	卢熠婷, 黄云飞, 陈刚. 一种蓄电池健康状态评估方法及装置: CN111142038A[P]. 2020-05-12.
	LU Y T, HUANG Y F, CHEN G. Storage battery health state evaluation method and device: CN111142038A[P]. 2020-05-12.
25	赵光财, 林名强, 戴厚德, 等. 一种锂电池SOH估计的KNN-马尔科夫修正策略[J]. 自动化学报, 2021, 47(2): 453-463.
	ZHAO G C, LIN M Q, DAI H D, et al. A modified strategy using the KNN-Markov chain for SOH estimation of lithium batteries[J]. Acta Automatica Sinica, 2021, 47(2): 453-463.
26	DRAGOMIRETSKIY K, ZOSSO D. Variational mode decomposition[J]. IEEE Transactions on Signal Processing, 2014, 62(3): 531-544.
27	李晓利, 高金峰. 用于配电网多目标无功优化的改进粒子群优化算法[J]. 电力自动化设备, 2019, 39(1): 106-111.
	LI X L, GAO J F. Improved particle swarm optimization algorithm for multi-objective reactive power optimization of distribution network[J]. Electric Power Automation Equipment, 2019, 39(1): 106-111.

电池型号	MAE	RMSE	MAPE/%	RA	RUL_e
B0005	0.0118	0.0148	0.83	0.9914	2
B0006	0.0091	0.0121	0.71	0.9934	1
B0007	0.0130	0.0137	0.88	0.9914	2
B0018	0.0082	0.0098	0.58	0.9942	1

电池型号	MAE	RMSE	MAPE/%	RA	RUL_e
B0005	0.0150	0.0177	1.07	0.9895	2
B0006	0.0150	0.0170	1.13	0.9891	3
B0007	0.0166	0.0192	1.11	0.9893	3
B0018	0.0151	0.0167	1.06	0.9895	2

电池型号	模型	MAE	RMSE	MAPE/%	RA	RUL_e
B0005	GRU	0.0670	0.0707	4.76	0.9512	7
	EMD-GRU	0.0281	0.0346	1.96	0.9792	4
	VMD-GRU	0.0143	0.0189	0.99	0.9893	3
	VMD-PSO-GRU	0.0118	0.0148	0.83	0.9914	2
B0006	GRU	0.0555	0.0620	4.11	0.9574	6
	EMD-GRU	0.0353	0.0433	2.61	0.9728	4
	VMD-GRU	0.0199	0.0215	1.55	0.9854	3
	VMD-PSO-GRU	0.0091	0.0121	0.71	0.9934	1
B0007	GRU	0.0470	0.0539	3.09	0.9680	5
	EMD-GRU	0.0302	0.0356	1.99	0.9794	4
	VMD-GRU	0.0211	0.0189	2.58	0.9891	3
	VMD-PSO-GRU	0.0130	0.0137	0.88	0.9914	2
B0018	GRU	0.0706	0.0805	4.96	0.9494	7
	EMD-GRU	0.0327	0.0381	2.30	0.9766	4
	VMD-GRU	0.0232	0.0242	1.65	0.9836	3
	VMD-PSO-GRU	0.0082	0.0098	0.58	0.9942	1

电池型号	模型	MAE	RMSE	MAPE	RUL_e
B0006	VMD-LSTM^[22]	0.0127	0.0312	0.0094	—
	ABMS-CEEMDAN-LSTM^[21]	0.0240	0.0308	—	3
	AAFSA-ELM^[10]	0.0157	0.0231	—	2
	VMD-PSO-GRU	0.0091	0.0121	0.0071	1

[1]	陈清炀, 何映晖, 余官定, 刘铭扬, 徐翀, 李振明. 模型与数据双驱动的锂电池状态精准估计[J]. 储能科学与技术, 2023, 12(1): 209-217.
[2]	田孟羽, 武怿达, 郝峻丰, 朱璟, 岑官骏, 乔荣涵, 申晓宇, 季洪祥, 金周, 詹元杰, 闫勇, 贲留斌, 俞海龙, 刘燕燕, 黄学杰. 锂电池百篇论文点评（2022.10.01—2022.11.30）[J]. 储能科学与技术, 2023, 12(1): 1-15.
[3]	沈馨, 张睿, 赵辰孜, 武鹏, 张羽彤, 张俊东, 范丽珍, 刘全兵, 陈爱兵, 张强. 金属锂电池中力-电化学机制研究进展[J]. 储能科学与技术, 2022, 11(9): 2781-2797.
[4]	吴敬华, 杨菁, 刘高瞻, 王脂胭, 张秩华, 俞海龙, 姚霞银, 黄学杰. 固态锂电池十年（2011—2021）回顾与展望[J]. 储能科学与技术, 2022, 11(9): 2713-2745.
[5]	陈淼淼, 邵钦君, 陈剑. 锂电池高比能量正极材料Cr₈O₂₁ 的制备及应用[J]. 储能科学与技术, 2022, 11(9): 3011-3020.
[6]	朱璟, 武怿达, 郝峻丰, 岑官骏, 乔荣涵, 申晓宇, 田孟羽, 季洪祥, 金周, 詹元杰, 闫勇, 贲留斌, 俞海龙, 刘燕燕, 黄学杰. 锂电池百篇论文点评（2022.6.1—2022.7.31）[J]. 储能科学与技术, 2022, 11(9): 3035-3050.
[7]	翟朋博, 常冬梅, 毕志杰, 赵宁, 郭向欣. 锂镧锆氧（LLZO）基固态锂电池界面关键问题研究进展[J]. 储能科学与技术, 2022, 11(9): 2847-2865.
[8]	尹涛, 贾隆舟, 常修亮, 戴作强, 郑莉莉. 软包磷酸铁锂电池高电压浮充后热安全研究[J]. 储能科学与技术, 2022, 11(8): 2546-2555.
[9]	石爽, 吕娜伟, 马敬轩, 尹康涌, 孙磊, 张宁, 金阳. 不同类型气体探测对磷酸铁锂电池储能舱过充安全预警有效性对比[J]. 储能科学与技术, 2022, 11(8): 2452-2462.
[10]	蔡兴初, 朱一鸣, 姜可尚, 席旭峰, 张艺超, 林惟实. 全氟己酮气体灭火系统在磷酸铁锂电池储能预制舱的应用[J]. 储能科学与技术, 2022, 11(8): 2497-2504.
[11]	陆剑心, 张英, 马出原, 邓康, 雷春英. 水凝胶对磷酸铁锂电池灭火实验性能[J]. 储能科学与技术, 2022, 11(8): 2637-2644.
[12]	宋来丰, 梅文昕, 贾壮壮, 王青松. 绝热条件下280 Ah大型磷酸铁锂电池热失控特性分析[J]. 储能科学与技术, 2022, 11(8): 2411-2417.
[13]	钟健平, 费韬. 基于WOA-BPNN的锂电池极片涂布缺陷检测识别[J]. 储能科学与技术, 2022, 11(8): 2537-2545.
[14]	申晓宇, 岑官骏, 乔荣涵, 朱璟, 季洪祥, 田孟羽, 金周, 闫勇, 武怿达, 詹元杰, 俞海龙, 贲留斌, 刘燕燕, 黄学杰. 锂电池百篇论文点评（2022.4.1—2022.5.31）[J]. 储能科学与技术, 2022, 11(7): 2007-2022.
[15]	黄鹏, 聂枝根, 陈峥, 舒星, 沈世全, 杨继鹏, 申江卫. 基于优化Elman神经网络的锂电池容量预测[J]. 储能科学与技术, 2022, 11(7): 2282-2294.