基于自适应多层RLS的锂离子电池参数辨识

doi:10.19799/j.cnki.2095-4239.2023.0605

摘要/Abstract

摘要：

电池参数的准确辨识是电动汽车电池管理系统实现高精度状态估计的基础。针对遗忘因子递推最小二乘法（forgetting factor recursive least squares，FFRLS）辨识变化电池参数时精度不足的问题，本文提出以层数形式更新参数的自适应多层递推最小二乘（adaptive multi-layer recursive least squares，AMLRLS）电池在线参数辨识方法。AMLRLS算法以第L-1层辨识参数的电压误差作为第L层的目标值，递推分离出电压误差中的参数量，以所有层的参数量之和作为一个数据点的辨识结果，形成多层RLS结构更新参数。针对每一次辨识算法都计算至最大设置层的问题，设计层数选择器，将第一层FFRLS辨识结果的电压误差作为层数选择器的输入量，以电压误差大小自适应选择层数，减小计算量。搭建电池模型，仿真验证AMLRLS的参数跟踪能力。仿真结果表明，AMLRLS的参数误差比RLS最大降低了69%，比AFFRLS（adaptive forgetting factor recursive least squares）最大降低了46.5%。在实验验证中，AMLRLS在DST（dynamic stress test）工况下相较其他算法电压均方根误差和平均绝对误差最大降低了43.9%和32.1%，不同电流、不同温度和不同初始SOC条件下的实验结果验证了AMLRLS具有较强的适用性。最后，实验比较了各算法的计算时间，相较于未设置层数选择器的情况，AMLRLS在DST工况下计算时间缩短了37.4%，在FUDS下缩短了28.6%，减少了电池管理系统的计算负担。

关键词: 锂离子电池, 参数辨识, 最小二乘法, 等效电路模型

Abstract:

Accurate identification of battery parameters is the foundation for achieving high-precision state estimation in electric vehicle battery management systems. To address the issue of insufficient accuracy when identifying changing battery parameters using the forgetting factor recursive least square (FFRLS) method, this study proposes an adaptive multilayer recursive least squares (AMLRLS) online battery parameter identification method, which updates parameters hierarchically. The AMLRLS algorithm uses the voltage error of the identified parameters of the L-1 layer as the target value for the L-th layer. It recursively separates the parameter quantities from the voltage error and aggregates them from all layers to form the identification result for a single data point, creating a MLRLS structure. To address the problem of computing up to the maximum set layer in each identification step, a layer selector is designed. It takes the voltage error from the FFRLS identification result of the first layer as input and adaptively selects the number of layers based on the magnitude of the voltage error, reducing computational load. A battery model is constructed, and simulations are conducted to verify the parameter tracking capability of AMLRLS. Results demonstrate that AMLRLS reduces parameter errors by up to 69% compared to RLS and 46.5% compared to adaptive FFRLS. In experimental validation, AMLRLS considerably reduces the root mean square error and average absolute error of voltage by 43.9% and 32.1%, respectively, under dynamic stress test (DST) conditions compared to other algorithms. Results across different currents, temperatures, and the initial state of charge conditions validate the strong applicability of AMLRLS. Finally, the computation time of various algorithms is compared. AMLRLS reduces computation time by 37.4% under DST conditions and by 28.6% under federal urban driving schedule conditions compared to scenarios without a layer selector, thus alleviating the computational burden of the battery management system.

Key words: lithium-ion battery, parameter identification, least squares, equivalent circuit model

中图分类号:

TM 912

段双明, 张胜利. 基于自适应多层RLS的锂离子电池参数辨识[J]. 储能科学与技术, 2024, 13(2): 712-720.

Shuangming DUAN, Shengli ZHANG. Lithium-ion battery parameter identification based on adaptive multilayer RLS[J]. Energy Storage Science and Technology, 2024, 13(2): 712-720.

图/表 12

图1

表1

AMLRLS参数辨识流程"

初始化 $θ ̂ L 1$ 至 $θ ̂ L m a x$ 、 $P$ 、 $e b a s e$ 、 $L m i n$ 、 $L m a x$ 、 $h$ 、 $λ$
步骤一：以式(11)更新 $e k$ ，以式(24)更新 $L k$
步骤二：以式(10)更新 $K (k)$ ，以式(13)更新 $P (k)$
步骤三：以式(12)更新 $θ ̂ L 1 (k)$ ，若 $L k = 1$ ，则进入步骤六
步骤四：以式(20)更新 $y L m (k)$ ，以式(21)更新 $e L m (k)$
步骤五：以式(22)更新 $θ ̂ L m (k)$ ，并判断是否计算到第 $L (k)$ 层。若是，则进入步骤六；若否，则以步骤四步骤五计算下一层参数
步骤六：以式(23)计算所有层的参数和 $θ ̂ (k)$

表1

图2

表2

图3

图4

图5

表3

图6

图7

表4

表5

参考文献 20

1	谭必蓉, 杜建华, 叶祥虎, 等. 基于模型的锂离子电池SOC估计方法综述[J]. 储能科学与技术, 2023, 12(6): 1995-2010.
	TAN B R, DU J H, YE X H, et al. Overview of SOC estimation methods for lithium-ion batteries based on model[J]. Energy Storage Science and Technology, 2023, 12(6): 1995-2010.
2	刘雨洋, 王顺利, 谢滟馨, 等. 基于在线参数辨识和改进2RC-PNGV模型的锂离子电池建模与SOC估算研究[J]. 储能科学与技术, 2021, 10(6): 2312-2317.
	LIU Y Y, WANG S L, XIE Y X, et al. Research on Li-ion battery modeling and SOC estimation based on online parameter identification and improved 2RC-PNGV model[J]. Energy Storage Science and Technology, 2021, 10(6): 2312-2317.
3	YANG R X, XIONG R, SHEN W X. On-board diagnosis of soft short circuit fault in lithium-ion battery packs for electric vehicles using an extended Kalman filter[J]. CSEE Journal of Power and Energy Systems, 2020, 8(1): 258-270.
4	FENG X N, PAN Y, HE X M, et al. Detecting the internal short circuit in large-format lithium-ion battery using model-based fault-diagnosis algorithm[J]. Journal of Energy Storage, 2018, 18: 26-39.
5	刘志聪, 张彦会. 锂离子电池参数辨识及荷电状态的估算[J]. 储能科学与技术, 2022, 11(11): 3613-3622.
	LIU Z C, ZHANG Y H. Parameter identification and state of charge estimation of lithium-ion batteries[J]. Energy Storage Science and Technology, 2022, 11(11): 3613-3622.
6	WANG S L, TAKYI-ANINAKWA P, FAN Y C, et al. A novel feedback correction-adaptive Kalman filtering method for the whole-life-cycle state of charge and closed-circuit voltage prediction of lithium-ion batteries based on the second-order electrical equivalent circuit model[J]. International Journal of Electrical Power & Energy Systems, 2022, 139: 108020.
7	SHI J J, GUO H S, CHEN D W. Parameter identification method for lithium-ion batteries based on recursive least square with sliding window difference forgetting factor[J]. Journal of Energy Storage, 2021, 44: 103485.
8	SHI H T, WANG S L, WANG L P, et al. On-line adaptive asynchronous parameter identification of lumped electrical characteristic model for vehicle lithium-ion battery considering multi-time scale effects[J]. Journal of Power Sources, 2022, 517: 230725.
9	ZHOU S D, LIU X H, HUA Y, et al. Adaptive model parameter identification for lithium-ion batteries based on improved coupling hybrid adaptive particle swarm optimization- simulated annealing method[J]. Journal of Power Sources, 2021, 482: 228951.
10	LEBEL F A, MESSIER P, SARI A, et al. Lithium-ioncell equivalent circuit model identification by galvano-static intermittent titration technique[J]. Journalof Energy Storage, 2022, 54: 105303.
11	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.
12	WANG S L, TAKYI-ANINAKWA P, JIN S Y, et al. An improved feedforward-long short-term memory modeling method for the whole-life-cycle state of charge prediction of lithium-ion batteries considering current-voltage-temperature variation[J]. Energy, 2022, 254: 124224.
13	TOWLIAT M, GUO Z, CIMINI L J, et al. Multi-layered recursive least squares for time-varying system identification[J]. IEEE Transactions on Signal Processing, 2022, 70: 2280-2292.
14	DU X H, MENG J H, ZHANG Y M, et al. An information appraisal procedure: Endows reliable online parameter identification to lithium-ion battery model[J]. IEEE Transactions on Industrial Electronics, 2022, 69(6): 5889-5899.
15	刘伟, 杨耕, 孟德越, 等. 计及常用恒流工况的锂离子电池建模方法[J]. 电工技术学报, 2021, 36(24): 5186-5200.
	LIU W, YANG G, MENG D Y, et al. Modeling method of lithium-ion battery considering commonly used constant current conditions[J]. Transactions of China Electrotechnical Society, 2021, 36(24): 5186-5200.
16	卫志农, 原康康, 成乐祥, 等. 基于多新息最小二乘算法的锂电池参数辨识[J]. 电力系统自动化, 2019, 43(15): 139-145.
	WEI Z N, YUAN K K, CHENG L X, et al. Parameter identification of lithium-ion battery based on multi-innovation least squares algorithm[J]. Automation of Electric Power Systems, 2019, 43(15): 139-145.
17	朱瑞, 段彬, 温法政, 等. 基于分布式最小二乘法的锂离子电池建模及参数辨识[J]. 机械工程学报, 2019, 55(20): 85-93.
	ZHU R, DUAN B, WEN F Z, et al. Lithium-ion battery modeling and parameter identification based on decentralized least squares method[J]. Journal of Mechanical Engineering, 2019, 55(20): 85-93.
18	SIMON H.自适应滤波器原理(第五版)[M]. 郑宝玉,等,译. 北京: 电子工业出版社. 2016: 1-253.
19	TRAN M K, MATHEW M, JANHUNEN S, et al. A comprehensive equivalent circuit model for lithium-ion batteries, incorporating the effects of state of health, state of charge, and temperature on model parameters[J]. Journal of Energy Storage, 2021, 43: 103252.
20	SIVA SURIYA NARAYANAN S, THANGAVEL S. Machine learning-based model development for battery state of charge-open circuit voltage relationship using regression techniques[J]. Journal of Energy Storage, 2022, 49: 104098.

项目	AMLRLS	AFFRLS	RLS
图2(a)MAE/F	8.1929	14.7935	21.8266
图2(b)MAE/F	11.4883	21.4742	37.0916

电流	方法	RMSE×10^-4/V	MAE×10^-4/V
DST	AMLRLS	8.7832	3.3571
	AFFRLS	14.1702	4.4647
	RLS	15.6808	4.9464

FUDS	AMLRLS	17.7456	6.0669
	AFFRLS	22.3721	7.2088
	RLS	24.5875	7.8741

电流文件	温度/℃	初始SOC/%	算法	RMSE×10^-4/V	MAE×10^-4/V
DST	0	50	AMLRLS	26.4872	9.8814
			AFFRLS	41.1485	14.9455
			RLS	44.8108	17.2250
		80	AMLRLS	21.7088	7.3372
			AFFRLS	33.9587	11.0849
			RLS	36.4855	12.4792
	45	50	AMLRLS	8.6529	3.1361
			AFFRLS	13.2339	4.8080
			RLS	14.4473	4.8633
		80	AMLRLS	4.9829	2.5069
			AFFRLS	6.7103	3.2715
			RLS	8.2974	3.4952
FUDS	0	50	AMLRLS	89.1604	46.0336
			AFFRLS	104.4175	54.9780
			RLS	112.4889	59.5618
		80	AMLRLS	50.2489	22.7675
			AFFRLS	59.5605	27.1466
			RLS	64.0396	29.1888
	45	50	AMLRLS	10.3818	4.6798
			AFFRLS	13.5460	5.8574
			RLS	16.4590	6.8622
		80	AMLRLS	7.0688	4.1482
			AFFRLS	8.6230	4.8412
			RLS	10.4544	5.2304

项目	DST工况辨识时间/s	FUDS工况辨识时间/s
AMLRLS	0.0839	0.0958
AMLRLS(未设置层数选择器)	0.1340	0.1341
AFFRLS	0.0521	0.0532
RLS	0.0507	0.0494

[1]	李校磊, 高健, 周伟东, 李泓. COMSOL Multiphysics在锂离子电池中的应用[J]. 储能科学与技术, 2024, 13(2): 546-567.
[2]	彭可, 张志成, 胡有章, 张旭辉, 周稼辉, 李彬. 基于有限元的热力耦合场匣钵运动分析与优化[J]. 储能科学与技术, 2024, 13(2): 634-642.
[3]	雷旗开, 余胤, 彭鹏, 陈满, 金凯强, 王青松. 隔热材料布局方式对280 Ah磷酸铁锂电池热失控传播抑制效果的影响[J]. 储能科学与技术, 2024, 13(2): 495-502.
[4]	李珂, 郝奕帆, 方振华, 王静, 张松通, 祝夏雨, 邱景义, 明海. 高功率化学电源体系发展及军事应用分析[J]. 储能科学与技术, 2024, 13(2): 436-461.
[5]	宋元明, 刘亚杰, 金光, 周星, 黄旭程. 锂离子电池/超级电容器混合储能系统能量管理方法综述[J]. 储能科学与技术, 2024, 13(2): 652-668.
[6]	杜文, 王君雷, 徐运飞, 李世龙, 王昆. 火焰喷雾热解法生产锂离子电池高镍三元正极材料的技术经济分析[J]. 储能科学与技术, 2024, 13(1): 345-357.
[7]	梁宏毅, 陈锋, 甘友毅, 邵丹. 动力锂电池三元正极低温性能研究[J]. 储能科学与技术, 2024, 13(1): 293-298.
[8]	肖也, 徐磊, 闫崇, 黄佳琦. 锂电池用参比电极的设计与应用[J]. 储能科学与技术, 2024, 13(1): 82-91.
[9]	戴雪娇, 闫婕, 王管, 董浩天, 蒋丹枫, 魏泽威, 孟凡星, 刘松涛, 张海涛. 铌基低温电池关键材料研究进展[J]. 储能科学与技术, 2024, 13(1): 311-324.
[10]	徐铖杰, 黄玉林, 林中飞雨, 林志铭, 方辰希, 张卫军, 黄志高, 李加新. 纳米硅的砂磨宏量制备及其碳纤维复合负极的储锂性能研究[J]. 储能科学与技术, 2024, 13(1): 1-11.
[11]	张怡, 葛筱渔, 李真, 黄云辉. 用于锂电池监测的声学和光学传感技术研究进展[J]. 储能科学与技术, 2024, 13(1): 167-177.
[12]	童文欣, 黄中垣, 王睿, 邓司浩, 何伦华, 肖荫果. 基于空间分辨中子衍射方法的锂离子电池电化学反应均匀性研究[J]. 储能科学与技术, 2024, 13(1): 72-81.
[13]	廖雅贇, 周峰, 张颖曦, 吕途安, 何阳, 陈晓燕, 霍开富. 锂离子电池快充石墨负极材料研究进展[J]. 储能科学与技术, 2024, 13(1): 130-142.
[14]	靳欣, 张建茹, 王其钰, 张锐, 王碧童, 张中洋, 俞海龙, 禹习谦, 李泓. 混合固液锂离子电池的热失控行为研究[J]. 储能科学与技术, 2024, 13(1): 48-56.
[15]	李召阳, 刘定宏, 赵岩岩, 陈满, 雷旗开, 彭鹏, 刘磊. 高比能量锂离子软包电池针刺测试的影响因素研究[J]. 储能科学与技术, 2024, 13(1): 57-71.