基于长短时记忆神经网络的锂离子电池多维老化诊断

doi:10.13229/j.cnki.jdxbgxb.20230056

摘要/Abstract

摘要：

本文通过引入两种对锂离子电池老化模式影响最大的内部副反应，改进传统伪二维模型的负极过电位方程，拓展建立锂离子电池性能衰退的电化学机理模型。应用响应面分析法，提取能够全面描述电池性能衰退的老化特征参数簇。建立一种长短时记忆神经网络，以基于机理模型获取的老化特征参数和历史容量保持率作为输入，预测电池未来容量衰退轨迹。结果表明：电池容量预测误差小于2%。

关键词: 锂离子电池, 老化诊断, 老化机理建模, 响应面分析法, 长短时记忆神经网络

Abstract:

Two internal side reactions that have the greatest impact on the battery aging mode are introduced. The negative region equation of the traditional pseudo two-dimensional model is improved， and the electrochemical degradation model of lithium-ion batteries is proposed. The response surface analysis method is applied to establish the aging characteristic parameters that can comprehensively describe the degradation of battery performance. A long short-term memory neural network is established to predict the future capacity. The aging characteristic parameters obtained based on the mechanism model and historical capacity retention rate are as the input of the network. Verification results of capacity forecast show that the prediction error is within 2%.

Key words: lithium-ion battery, aging diagnosis, aging mechanism modeling, response surface methodology, long short-term memory neural network

中图分类号:

TM912

任宪丰,袁文文,吴学强,时艳茹,姚蒙蒙,张凯旋,杨瑞鑫,潘悦. 基于长短时记忆神经网络的锂离子电池多维老化诊断[J]. 吉林大学学报(工学版), 2024, 54(11): 3135-3147.

Xian-feng REN,Wen-wen YUAN,Xue-qiang WU,Yan-ru SHI,Meng-meng YAO,Kai-xuan ZHANG,Rui-xin YANG,Yue PAN. Multi-dimensional aging diagnosis of lithium-ion battery with a long short-term memory neural network[J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(11): 3135-3147.

图/表 16

表1

图1

图2

表2

输入变量及其取值范围"

输入变量	Level Ⅰ	Level Ⅱ	Level Ⅲ
电解液锂离子浓度 c_l /（mol·m^-3）	1 200	1 120	1 080
SEI膜厚度 $δ f l i m$ /nm	1	400	650
析锂过电位η_Li/V	0.45	0.15	-0.2
负极可用锂离子浓度 c_s/（mol·m^-3）	30 665	30 392	30 242
负极电化学反应速率 k_n/［m^2.5·（mol^0.5·s）^-1］	4.38e-11	6.7e-11	1e-10
正极电化学反应速率 k_p/［m^2.5·（mol^0.5·s）^-1］	1.63e-11	2.5e-10	8e-10

表2

表3

表4

图3

图4

图5

图6

图7

图8

图9

图10

图11

表5

参考文献 34

1	Xiong R, Huang J T, Duan Y Z, et al. Enhanced lithium-ion battery model considering critical surface charge behavior[J]. Applied Energy, 2022, 314: No.118915.
2	Xiong R, Pan Y, Shen W X, et al. Lithium-ion battery aging mechanisms and diagnosis method for automotive applications: recent advances and perspectives[J]. Renewable and Sustainable Energy Reviews, 2020, 131: No.110048.
3	Xiong R, Ma S X, Li H L, et al. Toward a safer battery management system: a critical review on diagnosis and prognosis of battery short circuit[J]. Iscience, 2020, 23(4): No.101010.
4	Waldmann T, Wilka M, Kasper M, et al. Temperature dependent ageing mechanisms in lithium-ion batteries-a post-mortem study[J]. Journal of Power Sources, 2014, 262: 129-135.
5	Waldmann T, Iturrondobeitia A, Kasper M, et al. Post-mortem analysis of aged lithium-ion batteries: disassembly methodology and physico-chemical analysis techniques[J]. Journal of The Electrochemical Society, 2016, 163(10): No.21211609.
6	Waldmann T, Ghanbari N, Kasper M, et al. Correlations between electrochemical data and results from post-mortem analysis of aged lithium-ion batteries[J]. Journal of the Electrochemical Society, 2015, 162(8): No.20411508.
7	Klett M, Eriksson R, Groot J, et al. Non-uniform aging of cycled commercial LiFePO₄∥graphite cylindrical cells revealed by post-mortem analysis[J]. Journal of Power Sources, 2014, 257: 126-137.
8	Golla S U, Zeibig D, Prickler L, et al. Characterization of degeneration phenomena in lithium-ion batteries by combined microscopic techniques[J]. Micron, 2018, 113: 10-19.
9	Buqa H, Wursig A, Vetter J, et al. SEI film formation on highly crystalline graphitic materials in lithium-ion batteries[J]. Journal of Power Sources, 2006, 153(2): 385-390.
10	Darma M S D, Lang M, Kleiner K, et al. The influence of cycling temperature and cycling rate on the phase specific degradation of a positive electrode in lithium ion batteries: a post mortem analysis[J]. Journal of Power Sources, 2016, 327: 714-725.
11	Stiasazny B, Ziegler J C, Kraub E E, et al. Electrochemical characterization and post-mortem analysis of aged LiMn₂O₄–Li (Ni_0.5Mn_0.3Co_0.2) O₂/graphite lithium-ion batteries part I: cycle aging[J]. Journal of Power Sources, 2014, 251: 439-450.
12	Badey Q, Cherouvrier G, Reynier Y, et al. Mechanisms and modeling of lithium-ion battery aging for a vehicle usage[J]. ECS Meeting Abstracts IOP Publishing, 2011, 2(15): 742.
13	Jaumann T, Balach J, Klose M, et al. SEI-component formation on sub 5 nm sized silicon nanoparticles in Li-ion batteries: the role of electrode preparation, FEC addition and binders[J]. Physical Chemistry Chemical Physics, 2015, 17(38): 24956-24967.
14	Liu S, Jiang J, Shi W, et al. State of charge and peak power estimation of NCM/Li₄Ti₅O₁₂ battery using ic curve for rail tractor application[C]∥ 2014 IEEE Conference and Expo Transportation Electrification Asia-Pacific (ITEC Asia-Pacific), Beijing, China, 2014: 1-3.
15	Pan B, Dong D, Wang J G, et al. Aging mechanism diagnosis of lithium ion battery by open circuit voltage analysis[J]. Electrochimica Acta, 2020, 362: No.137101.
16	He J T, Bian X L, Liu L C, et al. Comparative study of curve determination methods for incremental capacity analysis and state of health estimation of lithium-ion battery[J]. Journal of Energy Storage, 2020, 29: No. 101400.
17	Li X Y, Yuan C G, Li X H, et al. State of health estimation for Li-ion battery using incremental capacity analysis and Gaussian process regression[J]. Energy, 2020, 190: No.116467.
18	Pan W J, Luo X S, Zhu M T, et al. A health indicator extraction and optimization for capacity estimation of Li-ion battery using incremental capacity curves[J]. Journal of Energy Storage, 2021, 42: No.103072.
19	Pastor F C, Widanage W D, Marco J, et al. Identification and quantification of ageing mechanisms in lithium-ion batteries using the EIS technique[C]∥ 2016 IEEE Transportation Electrification Conference and Expo (ITEC), Derborn, USA, 2016: 1-6.
20	Han X B, Ouyang M G, Lu L G, et al. A comparative study of commercial lithium ion battery cycle life in electrical vehicle: aging mechanism identification[J]. Journal of Power Sources, 2014, 251: 38-54.
21	Zhang Q, White R E. Capacity fade analysis of a lithium ion cell[J]. Journal of Power Sources, 2008, 179(2): 793-798.
22	Schmidt A P, Bitzer M, Imre Á W, et al. Model-based distinction and quantification of capacity loss and rate capability fade in Li-ion batteries[J]. Journal of Power Sources, 2010, 195(22): 7634-7638.
23	Ramadesigan V, Chen K, Burns N A, et al. Parameter estimation and capacity fade analysis of lithium-ion batteries using reformulated models[J]. Journal of the Electrochemical Society, 2011, 158(9):No. 13609926.
24	Fu R, Choe S Y, Agubra V, et al. Modeling of degradation effects considering side reactions for a pouch type Li-ion polymer battery with carbon anode[J]. Journal of Power Sources, 2014, 261: 120-135.
25	Wu J, Chen J X, Feng X, et al. State of health estimation of lithium-ion batteries using autoencoders and ensemble learning[J]. Journal of Energy Storage, 2022, 55: No.105708.
26	Yang B W, Wang D F, Zhang B, et al. Aging diagnosis-oriented three-scale impedance model of lithium-ion battery inspired by and reflecting morphological evolution[J]. Journal of Energy Storage, 2023, 59: No.106357.
27	Lee Y K. Effect of transition metal ions on solid electrolyte interphase layer on the graphite electrode in lithium ion battery[J]. Journal of Power Sources, 2021, 484: No.229270.
28	Chouchaine M, Arcelus O, Franco A. Heterogeneous solid-electrolyte interphase in graphite electrodes assessed by 4D-resolved computational simulations[J]. Batteries & Supercaps, 2021, 4(9): 1457-1463.
29	Ren D S, Smith K, Guo D X, et al. Investigation of lithium plating-stripping process in Li-ion batteries at low temperature using an electrochemical model[J]. Journal of the Electrochemical Society, 2018, 165(10): No.20661810.
30	Hein S, Danner T, Latz A. An electrochemical model of lithium plating and stripping in lithium ion batteries[J]. ACS Applied Energy Materials, 2020, 3(9): 8519-8531.
31	Doyle M, Newman J, Fuller T. Modeling of galvanostatic charge and discharge of the lithium/polymer/insertion cell[J]. Journal of the Electrochemical Society, 1993, 140(6): 1526-1533.
32	刘兴涛, 刘晓剑, 武骥, 等. 基于曲线压缩和极限梯度提升算法的锂离子电池健康状态估计[J]. 吉林大学学报: 工学版, 2022, 52(6): 1273-1280.
	Liu Xing-tao, Liu Xiao-jian, Wu Ji, et al. State of health estimation method for lithium⁃ion battery based on curve compression and extreme gradient boosting [J]. Journal of Jilin University (Engineering and Technology Edition), 2022, 52(6): 1273-1280.
33	高金武, 贾志桓, 王向阳,等. 基于PSO-LSTM的质子交换膜燃料电池退化趋势预测[J]. 吉林大学学报: 工学版, 2022, 52(9): 2192-2202.
	Gao Jin-wu, Jia Zhi-huan, Wang Xiang-yang, et al. Degradation trend prediction of proton exchange membrane fuel cell based on PSO⁃LSTM[J]. Journal of Jilin University (Engineering and Technology Edition), 2022, 52(9): 2192-2202.
34	Sak H, Senior A W, Beaufavs F. Long short-term memory recurrent neural network architectures for large scale acoustic modeling[J]. Interspeech, 2014, 2: 338-342.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

多项式各项	各项系数	多项式各项	各项系数
A	-5.79e11	A×B	0
B	-3.075e10	A×C	-7.4e7
C	2.07	A×D	-4.11e10
D	-61.78	A×E	-6.07e-8
E	-0.46	A×F	2.1e7
F	-2.77	B×C	5.651e7
A²	0	B×D	-2.08e10
B²	0	B×E	5.23e6
C²	-5.06e-4	B×F	-2.29e-5
D²	-21.746	C×D	0.019
E²	3.87e-5	C×E	-1.93e-5
F²	4.59e-5	C×F	-2.96e-5

方差来源	平方和	自由度	均方和	F值	p值
A	15.04	1	15.04	6.72	0.015 4
B	35.04	1	35.04	15.67	0.000 5
C	3.375	1	3.375	1.51	0.230 3
D	92.04	1	92.04	41.18	<0.000 1
E	360.38	1	360.375	161.11	<0.000 1
F	0.17	1	0.167	0.075	0.787 0
A×B	2.00	1	2.00	0.90	0.353 1
A×C	0.125	1	0.125	0.06	0.814 9
A×D	2.25	1	2.25	1.006	0.325 1
A×E	0	1	0	0	1
A×F	0.125	1	0.125	0.056	0.814 9
B×C	0.125	1	0.125	0.056	0.814 9
B×D	0.45	1	0.50	0.224	0.640 3
B×E	0.062 5	1	0.062 5	0.028	0.868 5
B×F	0	1	0	0	1
C×D	1.125	1	1.125	0.503	0.484 5
C×E	1.125	1	1.125	0.503	0.484 5
C×F	2.25	1	2.25	1.006	0.325 1
D×E	0	1	0	0	1
D×F	0.125	1	0.125	0.056	0.815 0
E×F	8	1	8	3.576	0.069 8
A²	3.17	1	3.17	1.42	0.244 3
B²	94.29	1	94.29	42.15	<0.000 1
C²	23.57	1	23.57	10.54	0.003 2
D²	24.89	1	24.89	11.13	0.002 6
E²	94.29	1	94.29	42.15	<0.000 1
F²	57.34	1	57.34	25.63	<0.000 1
模型	58.16	26	2.24
失拟项	55.77	21	2.66
纯误差	2.39	5	0.48
和	1 086.98	53

评价指标	方法	考虑老化特征参数体系	不考虑老化特征参数体系
MAE	LSTM RNN	0.006 3	0.011 4
MAE	BP	0.008 9	0.014 2
RMSE	LSTM RNN	0.009 2	0.014 2
RMSE	BP	0.011 6	0.016 5
NSE	LSTM RNN	0.999 3	0.913 1
NSE	BP	0.981 2	0.942 7

[1]	李房云,夏容,张怡欣. 考虑电池荷电状态的混合动力汽车复合电源协同控制[J]. 吉林大学学报(工学版), 2024, 54(4): 1114-1119.
[2]	潘凤文,弓栋梁,高莹,徐明伟,麻斌. 基于锂离子电池线性化模型的电流传感器故障诊断[J]. 吉林大学学报(工学版), 2021, 51(2): 435-441.
[3]	龚敏明, 时玮, 张言茹, 姜君, 张维戈, 姜久春. 纯电动公交车锂离子动力电池的使用条件控制[J]. 吉林大学学报(工学版), 2014, 44(4): 1081-1087.
[4]	张彩萍, 姜久春. 用基于遗传优化的扩展卡尔曼滤波算法辨识电池模型参数 [J]. , 2012, (03): 732-737.
[5]	李杨, 王永武, 张火成, 张娜, 邹玉峰. 磷酸铁锂动力电池动态SOC状态下的分选[J]. 吉林大学学报(工学版), 2011, 41(增刊2): 331-333.
[6]	刘险峰，邹积岩，李立伟 . 大容量蓄电池组的连接方式[J]. 吉林大学学报(工学版), 2007, 37(03): 672-0674.