SOC estimation method of power battery based on BP artificial neural network

doi:10.12028/j.issn.2095-4239.2019.0027

Abstract

Abstract: The state of charge (SOC) is one of the key technologies in battery management system. Among many estimation methods, neural network has obvious advantages in accuracy and robustness of estimation. The huge amount of data is an important factor to obtain the accurate value of SOC. To solve the above problems, a method of estimating SOC of power batteries based on BP artificial neural network is proposed. Taking a certain type of batteries as the experimental object, a large amount of data is obtained by collecting the data of voltage, current, internal resistance and temperature of batteries. Equivalent circuit model of battery was established, and the initial data were corrected considering the effects of battery polarization, charge-discharge ratio and temperature. BP artificial neural network model is established based on MATLAB platform, and the data are modified to train the network model, and the feasibility of the model is verified. The model is applied to the prediction of experimental data, and the estimation of SOC is realized by function fitting. Finally, by comparing the predicted and measured values of SOC, the validity of the artificial neural network model for estimating SOC is proved.

Key words: power battery, equivalent circuit, data correction, neural network model, SOC estimation

CLC Number:

SU Zhenhao, LI Xiaojie, QIN Jin, DU Wenjie, HAN Ning. SOC estimation method of power battery based on BP artificial neural network[J]. Energy Storage Science and Technology, 2019, 8(5): 868-873.

References

[1] 潘海鸿, 吕治强, 李君子, 等. 基于灰色扩展卡尔曼滤波的锂离子电池荷电状态估算[J]. 电工技术学报, 2017, 32(21):1-8. PAN Haihong, LÜ Zhiqiang, LI Junzi, et al. Charge state estimation of lithium ion batteries based on grey extended Kalman filter[J]. Transactions of China Electrotechnical Society, 2017, 32(21):1-8.
[2] 封进. BP神经网络预估锂离子电池SOC训练数据选择[J]. 电源技术, 2016, 40(2):283-286. FENG Jin. Selection of SOC training data for lithium ion batteries predicted by neural network[J]. Chinese Journal of Power Sources, 2016, 40(2):283-286.
[3] ANTÓN J C, NIETO P J, DE COS JUEZ F J, et al. Battery state-ofcharge estimator using the MARS technique[J]. IEEE Transactions on Power Electronics, 2013, 28(8):3798-3805.
[4] 杜涛, 李爱魁, 马军, 等. 动力电池SOC预估方法研究进展[J]. 电源技术, 2015, 39(4):844-848. DU Tao, LI Aikui, MA Jun, et al. Research progress of SOC prediction methods for power batteries[J]. Chinese Journal of Power Sources, 2015, 39(4):844-848.
[5] 王琪, 孙玉坤, 倪福银, 等. 一种混合动力电动汽车电池荷电状态预测的新方法[J]. 电工技术学报, 2016, 31(9):189-196. WANG Qi, SUN Yukun, NI Fuyin, et al. A new method for predicting battery charging state of hybrid electric vehicle[J]. Transactions of China Electrotechnical Society, 2016, 31(9):189-196.
[6] 杨杰, 王婷, 杜春雨, 等. 锂离子电池模型研究综述[J]. 储能科学与技术, 2019, 8(1):58-64. YANG Jie, WANG Ting, DU Chunyu, et al. Lithium ion battery model research synthesis[J]. Energy Storage Science and Technology, 2019, 8(1):58-64.
[7] 嵇雷, RYAD Chellali. 基于自适应卡尔曼滤波的电池荷电状态估算[J]. 电池, 2018, 48(4):240-243. JI Lei, RYAD Chellali. Battery charge state estimation vased on adaptive Kalman filter[J]. Battery Bimonthly, 2018, 48(4):240-243.
[8] 梁奇. 基于无迹卡尔曼滤波的锂电池SOC估算[D]. 绵阳:西南科技大学, 2018. LIAGN Qi. Estimation of lithium battery SOC based on unscented Kalman filter[D]. Mianyang:Southwest University of Science and Technology, 2018.
[9] 赵佳美. 基于二阶EKF的锂离子电池SOC估计的建模与仿真[D]. 西安:西安科技大学, 2018. ZHAO Jiamei. Modeling and simulation of SOC estimation for lithium-ion batteries based on second-order EKF[D]. Xi'an:Xi'an University of Science and Technology, 2018.
[10] 杨阳, 殷时蓉, 张栋省. 基于改进EKF的锂离子电池SOC估算[J]. 电池, 2018, 48(3):182-184. YANG Yang, YIN Shirong, ZHANG Dongsheng. Estimation of SOC of lithium-ion batteries based on improved EKF[J]. Battery Bimonthly, 2018, 48(3):182-184.
[11] 刘艳莉, 戴胜, 程泽, 等. 基于有限差分扩展卡尔曼滤波的锂离子电池SOC估计[J]. 电工技术学报, 2014, 29(1):221-228. LIU Yanli, DAI Sheng, CHENG Ze, et al. SOC Estimation of lithium ion batteries based on finite difference extended Kalman filter[J]. Transactions of China Electrotechnical Society, 2014, 29(1):221-228.
[12] 李志鹏, 赵杨. 纯电动汽车电池管理系统及SOC精确估计[J]. 电源技术, 2016, 140(5):1090-1093. LI Zhipeng, ZHAO Yang. Battery management system of pure electric vehicle and accurate estimation of SOC[J]. Chinese Journal of Power Sources, 2016, 140(5):1090-1093.
[13] XIONG Rui, HE Hongwen, SUN Fengchun, et al. Evaluation on state of charge estimation of batteries with adaptive extended Kalman filter by experiment approach[J]. IEEE Transactions on Vehicular Technology, 2013, 62(1):108-117.
[14] 赵轩, 马建, 刘瑞, 等. 基于GGAP-RBF神经网络的多参数纯电动客车蓄电池荷电状态预测[J]. 中国公路学报, 2015, 28(4):116-126. ZHAO Xuan, MA Jian, LIU Rui, et al. Prediction of multi-parameter pure electric bus battery charging status based on GGAP-RBF neural network[J]. China Journal of Highway and Transport, 2015, 28(4):116-126.
[15] 谢长君, 费亚龙, 曾春年, 等. 基于无迹粒子滤波的车载锂离子电池状态估计[J]. 电工技术学报, 2018, 33(17):3958-3964. XIE Changjun, FEI Yalong, ZENG Chunnian, et al. State estimation of vehicle-borne lithium-ion batteries based on trackless particle filtering[J]. Transactions of China Electrotechnical Society, 2018, 33(17):3958-3964.
[16] 庞辉. 基于扩展单粒子模型的锂离子电池参数识别策略[J]. 物理学报, 2018, 67(5):259-269. PANG Hui. Lithium ion battery parameter identification strategy based on extended single event model[J]. Acta Physica Sinica, 2018, 67(5):259-269.
[17] 吕杨蒙, 朱自伟, 刘宝泉. 一种在线估算蓄电池内阻的计算方法[J]. 储能科学与技术, 2019, 8(2):264-268. LÜ Yangmeng, ZHU Ziwei, LIU Baoquan. An online calculation method for estimating the internal resistance of batteries[J]. Energy Storage Science and Technology, 2019, 8(2):264-268.
[18] 马龙, 文华. 锂离子电池温度变化过程仿真与验证[J]. 储能科学与技术, 2018, 7(4):712-717. MA Long, WEN Hua. Lithium ion battery temperature change simulation and verification[J]. Energy Storage Science and Technology, 2018, 7(4):712-717.
[19] 张廷, 胡社教. 基于改进Thenenin模型锂电池SOC估算方法[J]. 电源技术, 2015, 39(11):2400-2403. ZHANG Yan, HU Shejiao. SOC estimation method for lithium battery based on improved Thenenin model[J]. Chinese Journal of Power Sources, 2015, 39(11):2400-2403.