基于BPNN-EKF-GD-RF算法的锂离子电池组荷电状态估计方法

doi:10.3901/JME.2025.12.251

摘要/Abstract

摘要： 锂离子电池模组的荷电状态估计(State-of-charge,SOC)是影响电池性能的一个重要内部状态，是电池组进行其它状态估计的基础。然而它的估计准确性易受温度等外部因素影响，且电池间的不一致性也为电池组中各单体电池的SOC估计带来了困难。提出一种将BP神经网络(Back propagation neural network,BPNN)与扩展卡尔曼滤波(Extended Kalman filter,EKF)算法相结合的电池组SOC估计方法。该方法首先基于先验SOC利用BPNN估计不同温度下“领导者”电池的端电压，将其与实测端电压对比后采用EKF算法完成SOC后验估计，同时基于电压差采用梯度下降(Gradient descent,GD)算法更新BPNN的输出层权重使算法更快收敛。在此基础上，设计修正策略利用随机森林(Random forest,RF)算法对“跟随者”电池的SOC进行调整估计。试验结果表明，所提的BPNN-EKF-GD-RF算法能实现电池组在不同温度下SOC的准确估计，常温下SOC估计误差保持在2.5%以内，在温度变化下电池组中单体电池SOC估计最大误差不超过3.2%，为复杂环境下锂离子电池组的SOC估计提供了一种高精度低复杂度方案。

关键词: SOC估计, BP神经网络, 扩展卡尔曼滤波, 梯度下降算法, 随机森林, 锂离子电池组

Abstract: The state-of-charge(SOC) estimation of lithium-ion battery modules is recognized as a critical internal state that significantly impacts battery performance and serves as the foundation for other state estimations in battery packs. However, the accuracy of SOC estimation is easily influenced by external factors such as temperature, and the inconsistency among batteries poses additional challenges for SOC estimation of cells within the pack. A method combining the back propagation neural network(BPNN) and the extended Kalman filter(EKF) algorithm is proposed for SOC estimation of battery packs. In this method, the terminal voltage of the "leader" battery at different temperatures is first estimated using BPNN based on the prior SOC. The estimated voltage is then compared with the measured voltage, and the posterior SOC estimation is completed using the EKF algorithm. Simultaneously, the output layer weights of the BPNN are updated using the gradient descent(GD) algorithm based on the voltage difference to accelerate algorithm convergence. Furthermore, a correction strategy is designed, and the random forest(RF) algorithm is utilized to adjust the SOC estimation of the "follower" batteries. Experimental results demonstrate that the proposed BPNN-EKF-GD-RF algorithm achieves accurate SOC estimation for battery packs under varying temperatures, with the SOC estimation error remaining within 2.5% at room temperature and the maximum SOC estimation error for cells in the pack not exceeding 3.2% under temperature variations. The research results provide a high-precision and low-complexity solution for SOC estimation of lithium-ion battery packs in complex environments.

Key words: SOC estimation, BP neural network, extended Kalman filter, gradient descent algorithm, random forest algorithm, lithium-ion battery pack

中图分类号:

U469

来鑫, 翁嘉辉, 杨一鹏, 孙宇飞, 周龙, 郑岳久, 韩雪冰. 基于BPNN-EKF-GD-RF算法的锂离子电池组荷电状态估计方法[J]. 机械工程学报, 2025, 61(12): 251-265.

LAI Xin, WENG Jiahui, YANG Yipeng, SUN Yufei, ZHOU Long, ZHENG Yuejiu, HAN Xuebing. State of Charge Estimation of Lithium-ion Battery Pack Based on BPNN-EKF-GD-RF Algorithm[J]. Journal of Mechanical Engineering, 2025, 61(12): 251-265.

参考文献

[1] WANG G，KONG D，PING P，et al. Revealing particle venting of lithium-ion batteries during thermal runaway：A multi-scale model toward multiphase process[J]. eTransportation，2023，16：100237.
[2] 孙广明，陈良亮，王瑞升，等. 基于深度强化学习的充光储能源站调度策略[J]. 电力工程技术，2021，40(5)：17-24. SUN Guangming，CHEN Liangliang，WANG Ruisheng，et al. Scheduling strategy for photovoltaic energy storage stations based on deep reinforcement learning [J]. Electric Power Engineering Technology，2021，40(5)：17-24.
[3] 耿健，杨冬梅，高正平，等. 含储能的冷热电联供分布式综合能源微网优化运行[J]. 电力工程技术，2021，40(1)：25-32. GENG Jian，YANG Dongmei，GAO Zhengping，et al. Optimization operation of distributed integrated energy microgrid with energy storage for combined cooling，heating and power supply[J]. Electric Power Engineering Technology，2021，40(1)：25-32.
[4] STREB M，ANDERSSON M，LÖFQVIST KLASS V，et al. Investigating re-parametrization of electrochemical model-based battery management using real-world driving data[J]. eTransportation，2023，16：100231.
[5] 唐传雨，韩华春，史明明，等. 基于DEKF的储能电池系统SOC估计方法研究[J]. 电力工程技术，2021，40(3)：7-14. TANG Chuanyu，HAN Huachun，SHI Mingming，et al. Research on SOC estimation method of energy storage battery system based on DEKF[J]. Electric Power Engineering Technology，2021，40(3)：7-14.
[6] 缪家森，成丽珉，吕宏水. 基于PSO-ELM的储能锂电池荷电状态估算[J]. 电力工程技术，2020，39(1)：165-169. MIAO Jiasen，CHENG Limin，LÜ Hongshui. State of charge estimation of energy storage lithium batteries based on PSO-ELM[J]. Electric Power Engineering Technology，2020，39(1)：165-169.
[7] 刘海东，周萍，周正，等. 锂离子电池开路电压快速估计研究[J]. 机械工程学报，2022，58(8)：227-235，243. LIU Haidong，ZHOU Ping，ZHOU Zheng，et al. Fast estimation of open circuit voltage for lithium-ion batteries[J]. Journal of Mechanical Engineering，2022，58(8)：227-235，243.
[8] 廖友萍，李睿，吕航，等. 基于EKF的钛酸锂电池SOC估计[J]. 电源技术，2023，47(5)：639-643. LIAO Youping，LI Rui，LÜ Hang，et al. SOC estimation of lithium titanate battery based on EKF[J]. Chinese Journal of Power Sources，2023，47(5)：639-643.
[9] ZHOU X，LIU Y，CHANG P，et al. Voltage stability analysis of a power system with wind power based on the Thévenin equivalent analytical method[J]. Electronics，2022，11(11)：1758.
[10] JORGENSEN C H L，MOLLER J G，SOMMER S，et al. A memory-efficient parallelizable method for computation of Thévenin equivalents used in real-time stability assessment[J]. IEEE Transactions on Power Systems，2019，34(4)：2675-2684.
[11] LIANG Y，WANG S，FAN Y，et al. Joint state of charge and state of health estimation of lithium-ion battery using improved adaptive dual extended Kalman filter based on piecewise forgetting factor recursive least squares[C]//20224th International Conference on Smart Power & Internet Energy Systems (SPIES). 2022：1923-7.10.1109/spies55999.2022.10082478.
[12] JIANG C，WANG S，WU B，et al. A state-of-charge estimation method of the power lithium-ion battery in complex conditions based on adaptive square root extended Kalman filter[J]. Energy，2021，219：119603.
[13] 申江卫，周灿彪，舒星，等. 宽温度环境下基于改进电化学模型的锂电池荷电状态估计[J]. 储能科学与技术，2023，12(9)：2904-2916. SHEN Jiangwei，ZHOU Canbiao，SHU Xing，et al. State of charge estimation for lithium batteries based on an improved electrochemical model at wide temperature environment[J]. Energy Storage Science and Technology，2023,12(9)：2904-2916.
[14] GAO Y，LIU K，ZHU C，et al. Co-estimation of state-of-charge and state-of-health for lithium-ion batteries using an enhanced electrochemical model[J]. IEEE Transactions on Industrial Electronics，2022，69(3)：2684-2696.
[15] 朱文凯，周星，刘亚杰，等. 基于递推门控循环单元神经网络的锂离子电池荷电状态实时估计方法[J]. 储能科学与技术，2023，12(2)：570-578. ZHU Wenkai，ZHOU Xing，LIU Yajie，et al. Real time state of charge estimation method of lithium-ion battery based on recursive gated recurrent unit neural network[J]. Energy Storage Science and Technology，2023，12(2)：570-578.
[16] 陈继斌，李雯雯，孙彦玺，等. 基于卷积-双向长短期记忆网络的电池SOC预测[J]. 电源技术，2022，46(5)：532-535. CHEN Jibin，LI Wenwen，SUN Yanxi，et al. Battery SOC prediction based on convolution bidirectional long term and short term memory network[J]. Chinese Journal of Power Sources，2022，46(5)：532-535.
[17] 李长龙，崔纳新，常龙，等. 风冷并联电池模组的建模方法与不一致性研究[J]. 机械工程学报，2022，58(12)：168-179. LI ChangLong，CUI Naxin，CHANG Long，et al. Modeling and inconsistency analysis of parallel-connected battery module with forced-air cooling[J]. Journal of Mechanical Engineering，2022，58(12)：168-179.
[18] 王珊珊，段续皇，任晓平，等. 储能电池模组双向主动均衡系统设计[J]. 电源技术，2023，47(2)：256-259. WANG Shanshan，Duan Xuhuang，REN Xiaoping，et al. Design of bidirectional active equalization system for battery energy storage module[J]. Chinese Journal of Power Sources，2023，47(2)：256-259.
[19] 吴铁洲，李梓豪. 基于模糊算法的锂离子电池组主动均衡方法[J]. 电源技术，2022，46(10)：1208-1212. WU Tiezhou，LI Zihao，Active equalization method for lithium ion battery pack based on fuzzy algorithm[J]. Chinese Journal of Power Sources，2022，46(10)：1208-1212.
[20] 孙正，李军，李虎林. 考虑温度的电池荷电状态估算和主被动均衡[J]. 科学技术与工程，2022，22(33)：14767-14778. SUN Zheng，LI Jun，LI Hulin. Battery state estimation and active passive equalization considering temperature[J]. Science Technology and Engineering，2022，22(33)：14767-14778.
[21] 王鹿军，单恩泽. 基于动态式双阈值的锂电池组主被动均衡策略[J]. 电机与控制学报，2022，26(1)：126-136. WANG Lujun，SHAN Enze. Active passive equalization strategy for lithium battery pack based on dynamic dual threshold[J]. Electric Machines and Control，2022，26(1)：126-136.
[22] 侯亚虹. 动力电池组不一致性建模及SOC融合估计研究[D]. 西安：长安大学，2021. HOU Yahong. Inconsistency modeling of power battery pack and SOC fusion estimation[D]. Xi’an：Chang’an University，2021.
[23] 蒋帅，陈铖，段砚州，等. 温度和老化影响下的锂离子电池荷电状态和健康状态联合估计方法研究[J]. 机械工程学报，2024，60(18)：266-275. JIANG Shuai，CHEN Cheng，DUAN Yanzhou，et al. Joint estimation of state-of-charge and state-of-health for lithium-ion batteries under the influence of temperature and aging[J]. Journal of Mechanical Engineering，2024，60(18)：266-275.
[24] ZHENG Y，GAO W，OUYANG M，et al. State-of-charge inconsistency estimation of lithium-ion battery pack using mean-difference model and extended Kalman filter [J]. Journal of Power Sources，2018，383：50-58.
[25] 金建新，虞儒新，刘刚，等. 锂离子电池健康状态估算方法研究进展[J]. 电气工程学报，2024，19(1)：33-48. JIN Jianxin，YU Ruxin，LIU Gang，et al. Research progress on state-of-health estimating method for lithium-ion batteries[J]. Journal of Electrical Engineering，2024，19(1)：33-48.
[26] Lai X，Huang Y，Han X，et al. A novel method for state of energy estimation of lithium-ion batteries using particle filter and extended Kalman filter[J]. Journal of Energy Storage，2021，43：103269.
[27] 黄云峰. 复杂工况与多因素影响下锂离子电池能量状态估计方法的研究[D]. 上海：上海理工大学，2022. Huang Yunfeng. Research on state of energy estimation method of lithium-ion battery under the influence of multiple factors[D]. Shanghai：University of Shanghai for Science and Technology，2022.
[28] 周航，刘晓龙，张梦迪，等. 基于简单循环单元的储能锂离子电池SOC和SOH联合估计方法[J]. 电气工程学报，2023，18(3)：332-340. ZHOU Hang，LIU Xiaolong，ZHANG Mengdi，et al. Joint SOC and SOH estimation method for energy storage lithium-ion batteries based on simple recurrent unit[J]. Journal of Electrical Engineering，2023，18(3)：332-340.
[29] 石琴，蒋正信，刘翼闻，等. 基于分数阶模型的锂离子电池荷电状态估计[J]. 机械工程学报，2024，60(8)：224-232，244. SHI Qin，JIANG Zhengxin，LIU Yiwen，et al. Fractional order model-based estimation for state of charge in lithium-ion battery[J]. Journal of Mechanical Engineering，2024，60(8)：224-232，244.
[30] 韦守李，尹涛，贾隆舟，等. 软包磷酸铁锂电池高温浮充后的电化学性能[J]. 电气工程学报，2024，19(2)：297-306. WEI Shouli，YIN Tao，JIA Longzhou，et al. Electrochemical properties of soft-pack LiFePO₄ battery after high-temperature float charge[J]. Journal of Electrical Engineering，2024，19(2)：297-306.
[31] 刘素贞，袁路航，张闯，等. 基于超声时域特征及随机森林的磷酸铁锂电池荷电状态估计[J]. 电工技术学报，2022，37(22)：5872-5885. LIU Suzhen，YUAN Luhang，ZHANG Chuang，et al. State of charge estimation of LiFeO₄ Batteries based on time domain features of ultrasonic waves and random forest[J]. Transactions of China Electrotechnical Society，2022，37(22)：5872-5885.
[32] SHU X，CHEN Z，SHEN J，et al. State of charge estimation for lithium-ion battery based on hybrid compensation modeling and adaptive H-infinity filter[J]. IEEE Transactions on Transportation Electrification，2023，9(1)：945-957.
[33] 李成，陈球，于莹莹，等. 基于IWOA-LSSVM的锂离子电池RUL预测[J]. 电气工程学报，2024，19(3)：399-411. LI Cheng，CHEN Qiu，YU Yingying，et al. Remaining useful life prediction for lithium-ion batteries based on IWOA-LSSVM[J]. Journal of Electrical Engineering，2024，19(3)：399-411.
[34] YANG N，SONG Z，HOFMANN H，et al. Robuststate of health estimation of lithium-ion batteries using convolutional neural network and random forest[J]. Journal of Energy Storage，2022，48：103857.