Remaining useful life prediction of lithium-ion batteries based on VF-DW-DFN

doi:10.19799/j.cnki.2095-4239.2021.0665

Abstract

Abstract:

Lithium-ion battery is an important part of different energy storage systems and equipment. An accurate forecast of lithium-ion battery remaining usable life is critical for guaranteeing the dependability and safety of battery-related enterprises and facilities. In this study, a new data-driven method is proposed to enhance the low forecasting accuracy and poor generalization performance in the remaining useful life prediction of lithium-ion batteries. Poor performance is frequently the result of a single data-driven strategy and is caused by non-stationary, nonlinear dynamics in battery deterioration dynamics. The proposed method is based on variational filtering, data wrapping, and a deep fusion network. First, the random noise interference of the original battery degradation sequence is eliminated to obtain the relatively stable degradation characteristic data by using the variational filtering method. Then, using the optimum embedding approach to accomplish feature data wrapping and limit information loss, an unique deep fusion network is constructed to model non-linear battery deterioration data, detect the degradation pattern in the battery data, and realize the final remaining usable life prediction. Lastly, remaining useful life prediction experiments are conducted on the lithium cobalt oxide battery data set and the average root means a square error of prediction is 1.41%, and the average absolute error of remaining useful life of prediction is less than two cycles. The suggested technique provides strong generalization, high forecasting accuracy, and minimal prediction error and can successfully anticipate the deterioration process of lithium-ion batteries, according to experimental data.

Key words: remaining useful life prediction, lithium-ion battery, variational filtering, data wrapping, deep fusion network

CLC Number:

TM 911.3

Shunmin YI, Linbo XIE, Li PENG. Remaining useful life prediction of lithium-ion batteries based on VF-DW-DFN[J]. Energy Storage Science and Technology, 2022, 11(7): 2305-2315.

Figures/Tables 16

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Table 1

Fig. 7

Table 2

Table 3

Fig. 8

Fig. 9

Table 4

Table 5

Fig. 10

Table 6

References 32

1	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.
2	OMARIBA Z B, ZHANG L J, SUN D B. Review on health management system for lithium-ion batteries of electric vehicles[J]. Electronics, 2018, 7: 72.
3	袁烨, 张永, 丁汉. 工业人工智能的关键技术及其在预测性维护中的应用现状[J]. 自动化学报, 2020, 46(10): 2013-2030.
	YUAN Y, ZHANG Y, DING H. Research on key technology of industrial artificial intelligence and its application in predictive maintenance[J]. Acta Automatica Sinica, 2020, 46(10): 2013-2030.
4	ZHANG W J. A review of the electrochemical performance of alloy anodes for lithium-ion batteries[J]. Journal of Power Sources, 2011, 196(1): 13-24.
5	康鑫, 时玮, 陈洪涛. 基于锂离子电池简化电化学模型的参数辨识[J]. 储能科学与技术, 2020, 9(3): 969-978.
	KANG X, SHI W, CHEN H T. Parameter identification based on simplified electrochemical model of lithium ion battery[J]. Energy Storage Science and Technology, 2020, 9(3): 969-978.
6	朱奕楠, 吕桃林, 赵芝芸, 等. 基于并行卡尔曼滤波器的锂离子电池荷电状态估计[J]. 储能科学与技术, 2021, 10(6): 2352-2362.
	ZHU Y N, LÜ T L, ZHAO Z Y, et al. State of charge estimation of lithium ion battery based on parallel Kalman filter[J]. Energy Storage Science and Technology, 2021, 10(6): 2352-2362.
7	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.
8	CHANG Y, FANG H J, ZHANG Y. A new hybrid method for the prediction of the remaining useful life of a lithium-ion battery[J]. Applied Energy, 2017, 206: 1564-1578.
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	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.
11	SEVERSON K A, ATTIA P M, JIN N, et al. Data-driven prediction of battery cycle life before capacity degradation[J]. Nature Energy, 2019, 4(5): 383-391.
12	PATIL M A, TAGADE P, HARIHARAN K S, et al. A novel multistage Support Vector Machine based approach for Li ion battery remaining useful life estimation[J]. Applied Energy, 2015, 159: 285-297.
13	CHANG Y, FANG H J. A hybrid prognostic method for system degradation based on particle filter and relevance vector machine[J]. Reliability Engineering & System Safety, 2019, 186: 51-63.
14	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.
15	陈峥, 李磊磊, 舒星, 等. 基于特征处理与径向基神经网络的锂电池剩余容量估算方法[J]. 储能科学与技术, 2021, 10(1): 261-270.
	CHEN Z, LI L L, SHU X, et al. Efficient remaining capacity estimation method for LIB based on feature processing and the RBF neural network[J]. Energy Storage Science and Technology, 2021, 10(1): 261-270.
16	CHEN L, ZHANG Y, ZHENG Y, et al. Remaining useful life prediction of lithium-ion battery with optimal input sequence selection and error compensation[J]. Neurocomputing, 2020, 414: 245-254.
17	XUE Z W, ZHANG Y, CHENG C, et al. Remaining useful life prediction of lithium-ion batteries with adaptive unscented Kalman filter and optimized support vector regression[J]. Neurocomputing, 2020, 376: 95-102.
18	SATEESH BABU G, ZHAO P L, LI X L. Deep convolutional neural network based regression approach for estimation of remaining useful life[C]//Database Systems for Advanced Applications, 2016.
19	MA G J, ZHANG Y, CHENG C, et al. Remaining useful life prediction of lithium-ion batteries based on false nearest neighbors and a hybrid neural network[J]. Applied Energy, 2019, 253: 113626.
20	易灵芝, 张宗光, 范朝冬, 等. 基于EEMD-GSGRU的锂电池寿命预测[J]. 储能科学与技术, 2020, 9(5): 1566-1573.
	YI L Z, ZHANG Z G, FAN C D, et al. Life prediction of lithium battery based on EEMD-GSGRU[J]. Energy Storage Science and Technology, 2020, 9(5): 1566-1573.
21	QIN T C, ZENG S K, GUO J B, et al. A rest time-based prognostic framework for state of health estimation of lithium-ion batteries with regeneration phenomena[J]. Energies, 2016, 9(11): 896.
22	DRAGOMIRETSKIY K, ZOSSO D. Variational mode decomposition[J]. IEEE Transactions on Signal Processing, 2014, 62(3): 531-544.
23	BOYD S. Distributed optimization and statistical learning via the alternating direction method of multipliers[J]. Foundations and Trends® in Machine Learning, 2010, 3(1): 1-122.
24	CAO L Y. Practical method for determining the minimum embedding dimension of a scalar time series[J]. Physica D: Nonlinear Phenomena, 1997, 110(1/2): 43-50.
25	SHEN Z P, ZHANG Y M, LU J W, et al. SeriesNet: A generative time series forecasting model[C]//2018 International Joint Conference on Neural Networks (IJCNN). July 8-13, 2018, Rio de Janeiro, Brazil. IEEE, 2018: 1-8.
26	LEA C, FLYNN M D, VIDAL R, et al. Temporal convolutional networks for action segmentation and detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. July 21-26, 2017, Honolulu, HI, USA. IEEE, 2017: 1003-1012.
27	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. June 27-30, 2016, Las Vegas, NV, USA. IEEE, 2016: 770-778.
28	CHANG S Y, ZHANG Y, HAN W, et al. Dilated recurrent neural networks[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. New York, USA: Curran Associates Inc., 2017: 76-86.
29	LAI G K, CHANG W C, YANG Y M, et al. Modeling long- and short-term temporal patterns with deep neural networks[C]//SIGIR '18: The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval. 2018: 95-104.
30	SAHA B, GOEBEL K. Battery data set[R]. NASA Ames Prognostics Data Repository, 2007.
31	LIU D T, ZHOU J B, LIAO H T, et al. A health indicator extraction and optimization framework for lithium-ion battery degradation modeling and prognostics[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2015, 45(6): 915-928.
32	FAN J M, FAN J P, LIU F, et al. A novel machine learning method based approach for Li-ion battery prognostic and health management[J]. IEEE Access, 2019, 7: 160043-160061.

电池编号	EOM/Ah	T_EOM	EOL/Ah	T_EOL
B5	1.6	74	1.4	125
B6	1.6	62	1.4	109
B7	1.6	93	1.42	160

网络类型	网络参数	参数值
DFN	输入单元数	d
	卷积核个数	32/32
	卷积核大小	5/5
	卷积核步长	1/1
	循环隐藏单元数	64/64/64
	自回归阶次	5
	输出单元数	1
	学习率	0.001

电池编号	RUL_r	RUL_p	RMSE	MAE	AE
B5	51	52	0.0136	0.0101	1
B6	47	48	0.0203	0.0117	1
B7	67	69	0.0085	0.0063	2

电池编号	滑窗大小	RMSE	MAE	AE
B5	15	0.0136	0.0101	1
	5	0.0213	0.0187	5
	30	0.0340	0.0314	11
B6	13	0.0203	0.0117	1
	5	0.0265	0.0193	3
	30	0.0468	0.0413	9
B7	15	0.0085	0.0063	2
	5	0.0330	0.0313	—
	30	0.0432	0.0418	—

电池编号	方法	RMSE	MAE	AE
B5	VF-DW-DFN	0.0136	0.0101	1
	VF-DW-SVM	0.0350	0.0331	10
	VF-DW-LSTM	0.0341	0.0305	10
B6	VF-DW-DFN	0.0203	0.0117	1
	VF-DW-SVM	0.0612	0.0555	17
	VF-DW-LSTM	0.0607	0.0564	18
B7	VF-DW-DFN	0.0085	0.0063	2
	VF-DW-SVM	0.0194	0.0177	—
	VF-DW-LSTM	0.0262	0.0247	—