An improved battery state estimation based on support vector machine regression

doi:10.19799/j.cnki.2095-4239.2020.0076

Abstract

Abstract:

The state-of-charge (SOC) estimation of a lithium-ion battery is very important with respect to a battery management system (BMS). It is difficult to ensure SOC estimation accuracy because it cannot be measured directly. To improve the SOC estimation accuracy, a least squares support vector machine (LSSVM) is used to establish a relation among voltage, current, and SOC. Further, an improved LSSVM method for SOC estimation is proposed to reduce the SOC estimation accuracy because of the changing voltage and current. The voltage measurement, current measurement, and SOC estimation values of the previous time are considered to be the feedback quantities of the model, and the voltage and current values of the present time are considered to be the input quantities that can be used to estimate the current SOC. The experimental results show that the error of the proposed method is less than 1% when compared with LSSVM, verifying the effectiveness of the proposed method.

Key words: lithium ion battery, SOC, LSSVM, feedback

CLC Number:

TM 911

LI Jiabo, WEI Meng, LI Zhongyu, YE Min, JIAO Shengjie, XU Xinxin. An improved battery state estimation based on support vector machine regression[J]. Energy Storage Science and Technology, 2020, 9(4): 1200-1205.

Figures/Tables 13

Fig.1

Table 1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

Table 2

References 10

1	GUO Y, ZHAO Z S, HUANG L M. SOC estimation of lithium battery based on improved BP neural network[J]. Energy Procedia, 2017, 105:4153-4158.
2	Comparison between two nonlinear Kalman filters for reliable SOC estimation on a prototypal BMS[C]//Conference of the IEEE Industrial Electronics Society, IEEE, 2016.
3	MENG J, RICCO M, LUO G, et al. An overview and comparison of online implementable SOC estimation methods for lithium-ion battery[J]. IEEE Transactions on Industry Applications, 2018: 1583-1591.
4	Capacity-loss diagnostic and life-time prediction in lithium-ion batteries: Part 1. Development of a capacity-loss diagnostic method based on open-circuit voltage analysis[J]. Journal of Power Sources, 2016, 301: 187-193.
5	ZHAO T, PENG X, PENG Y, et al. Lithium-ion battery SOC estimation method with fusion improved Kalman filter algorithm[J]. Chinese Journal of Scientific Instrument, 2016, 37(7): 1441-1448.
6	YAN Q, WANG Y. Predicting for power battery SOC based on neural network[C]//Proceedings of the 36th China Control Conference， 2017.
7	刘晓悦, 杜晓. 基于神经网络与UKF结合的锂离子电池组SOC估算方法[J]. 现代电子技术, 2018, 41(2): 120-123, 129.
	LIU X Y, DU X. SOC estimation method of lithium ion battery based on neural network and UKF[J]. Modern Electronic Technology, 2018, 41(2): 120-123, 129.
8	于洋, 纪世忠, 魏克新. 基于LS-SVM算法动力电池SOC估计方法的研究[J]. 电源技术, 2012, 36(3): 349-351.
	YU Y, JI S Z, WEI K X. Research on SOC estimation method of power battery based on LS-SVM algorithm[J]. Chinese Journal of Power Sources, 2012, 36(3): 349-351.
9	王琪, 孙玉坤, 黄永红. 基于参数优化的最小二乘支持向量机HEV阀控铅酸蓄电池SOC预测[J]. 中南大学学报(自然科学版), 2015(1): 113-119.
	WANG Q, SUN Y K, HUANG Y H. SOC prediction of valve regulated lead acid battery based on least square support vector machine[J]. Journal of Central South University(Natural Science Edition), 2015(1): 113-119.
10	顾燕萍, 赵文杰, 吴占松. 最小二乘支持向量机的算法研究[J]. 清华大学学报(自然科学版), 2010(7): 1063-1066.
	GU Y P, ZHAO W J, WU Z S. Algorithm research of LSSVM[J]. Journal of Tsinghua University(Natural Science Edition), 2010(7): 1063-1066.

方法	RMSE/%	MAE/%
提出的模型	0.76	0.36
LSSVM	2.58	10.13
SVM	2. 79	11.10
NN	2.96	14.05

[1]	Haitao LI, Lingli KONG, Xin ZHANG, Chuanjun YU, Jiwei WANG, Lin XU. The effects of N/P design on the performances of Ni-rich NCM/Gr lithium ion battery [J]. Energy Storage Science and Technology, 2022, 11(7): 2040-2045.
[2]	Sida HUO, Wendong XUE, Xinli LI, Yong LI. Visualization analysis of composite electrolytes for lithium battery based on CiteSpace [J]. Energy Storage Science and Technology, 2022, 11(7): 2103-2113.
[3]	XIN Yaoda, LI Na, YANG Le, SONG Weili, SUN Lei, CHEN Haosen, FANG Daining. Integrated sensing technology for lithium ion battery [J]. Energy Storage Science and Technology, 2022, 11(6): 1834-1846.
[4]	Biao MA, Chunjing LIN, Lei LIU, Xiaole MA, Tianyi MA, Shiqiang LIU. Venting characteristics and flammability limit of thermal runaway gas of lithium ion battery [J]. Energy Storage Science and Technology, 2022, 11(5): 1592-1600.
[5]	Feng TIAN, Zhijiang CHENG, Handi YANG, Tianxiang YANG. Fault-tolerant control strategy for modular multi-level hybrid converter battery energy storage system [J]. Energy Storage Science and Technology, 2022, 11(5): 1583-1591.
[6]	Luyu GAN, Rusong CHEN, Hongyi PAN, Siyuan WU, Xiqian YU, Hong LI. Multiscale research strategy of lithium ion battery safety issue： Experimental and simulation methods [J]. Energy Storage Science and Technology, 2022, 11(3): 852-865.
[7]	Pengchao HUANG, Jiaqiang E. State estimation of lithium-ion battery based on dual adaptive Kalman filter [J]. Energy Storage Science and Technology, 2022, 11(2): 660-666.
[8]	Xiaozhi GAO, Lei WANG, Jin TIAN, Jialu LIU, Qinghua LIU. Research on hybrid energy storage power distribution strategy based on parameter optimization variational mode decomposition [J]. Energy Storage Science and Technology, 2022, 11(1): 147-155.
[9]	Shanshan MA, Tingting FANG, Liuqian YANG, Shuwan HU. Application of chromatography-mass spectrometry in study of lithium ion battery [J]. Energy Storage Science and Technology, 2022, 11(1): 60-65.
[10]	Jinhui GAO, Yinglong CHEN, Fanhui MENG, Meichao DING, Li WANG, Gang XU, Xiangming HE. Research on in-situ optical microscopic observation in lithium-ion batteries [J]. Energy Storage Science and Technology, 2022, 11(1): 53-59.
[11]	Aifen SUN, Na CHI. SOC estimation algorithm based on improved Gaussian process regression [J]. Energy Storage Science and Technology, 2022, 11(1): 253-257.
[12]	Xinyu CAO, Fei PENG, Liwei LI, Jianguang YIN. SOC estimation of lithium battery based on IBAS-NARX neural network model [J]. Energy Storage Science and Technology, 2021, 10(6): 2342-2351.
[13]	Yi'nan ZHU, Taolin LÜ, Zhiyun ZHAO, Wen YANG. State of charge estimation of lithium ion battery based on parallel Kalman filter [J]. Energy Storage Science and Technology, 2021, 10(6): 2352-2362.
[14]	Kehuan XIE, Chuanchang LI, Jian CHEN, Longhai YU, zhun TAN, Weihai QIN. Simulation model advances in vanadium redox flow battery energy storage and monitoring method for state of charge [J]. Energy Storage Science and Technology, 2021, 10(6): 2363-2372.
[15]	Zeng'ang JIA, Zhibin LING, Xuguang LI. Thermal characteristics of lithium-ion battery with sinusoidal charge and discharge pulsating current [J]. Energy Storage Science and Technology, 2021, 10(6): 2260-2268.