欢迎来到OAJRC平台！

期刊栏目

投稿

审稿

期刊目次

加入编委

期刊订阅

添加您的邮件地址以接收即将发行期刊数据：

当前位置：电气工程与自动化 > 2025年 4卷 (5)期

Open Access Article

Journal of Electrical Engineering and Automation. 2025; 4: (5) ; 132-140 ; DOI: 10.12208/j.jeea.20250195.

State of charge estimation of lithium-ion batteries incorporating a hybrid kernel extreme learning machine
基于混合核极限学习机的锂电池状态融合估计

作者： 王赛, 沈德智, 丁洁 *, 程艳云

南京邮电大学自动化学院江苏南京

*通讯作者：丁洁,单位：南京邮电大学自动化学院江苏南京；

江苏省研究生科研创新计划（项目编号：KYCX25_1174）

发布时间: 2025-05-22 总浏览量: 127

PDF 全文下载引用本文

摘要

在智能交通系统中，电动汽车的能源管理效率直接影响交通系统的可靠性与可持续性，而锂离子电池荷电状态（State of Charge, SOC）的精准估计是其重中之重。针对SOC的非线性和动态特征，本文提出了一种基于多源数据融合的电池荷电状态估计模型。在数据预处理阶段，首先利用自适应噪声完全集合经验模态分解和小波阈值降噪技术对输入的电压和电流信号进行分解和去噪处理，便于后续SOC估计；在特征提取阶段，采用卷积神经网络对去噪后的电压和电流数据进行特征提取，并将提取的电压和电流信号中的隐含特征与去噪后的电压电流数据一起作为混合核极限学习机的输入，利用其非线性拟合能力和泛化能力对SOC进行估计；最后将混合核极限学习机的估计值作为卡尔曼滤波器的测量值，对SOC进行实时估计，提高SOC估计的准确性以及模型的鲁棒性。基于McMaster大学提供的实验数据，对比其他模型，所提出的模型在SOC估计精度上表现出显著优势，具体而言，在25°C时均方根误差从1.95%降低到0.31%，平均绝对误差从2.11%降低到0.22%。

关键词： 荷电状态估计；混合核极限学习机；卷积神经网络；自适应噪声完全集合经验模态分解；小波阈值降噪；卡尔曼滤波

Abstract

In intelligent transportation systems, the energy management efficiency of electric vehicles directly affects the reliability and sustainability of the transportation system. Accurately estimating the State of Charge (SOC) of lithium-ion batteries is of paramount importance. Considering the nonlinear and dynamic characteristics of SOC, this paper proposes a battery SOC estimation model based on multi-source data fusion. During the data preprocessing stage, the input voltage and current signals are decomposed and denoised by complete ensemble empirical modal decomposition with adaptive noise and wavelet threshold denoising techniques to facilitate subsequent SOC estimation. In the feature extraction stage, a convolutional neural network is employed to extract features from the denoised voltage and current data. The extracted implicit features of the voltage and current signals, along with the denoised voltage and current data, are utilized as inputs to the hybrid kernel extreme learning machine, which leverages its nonlinear fitting and generalization capabilities to estimate the SOC. Finally, the estimated value from the hybrid kernel extreme learning machine is utilized as the measurement value for the kalman filter to perform real-time SOC estimation, enhancing the accuracy of SOC estimation and the robustness of the model. Based on experimental data provided by McMaster University, compared with other models, the proposed model demonstrates a significant advantage in SOC estimation accuracy. Specifically, the root mean square error is reduced from 1.95% to 0.31%, and the mean absolute error is reduced from 2.11% to 0.22% at 25°C.

Key words： State of charge; HKELM; CNN; Complete ensemble empirical mode decomposition with adaptive noise; Wavelet threshold denoising; Kalman filter

参考文献 References

[1] 安富强, 赵洪量, 程志, 等. 纯电动车用锂离子电池发展现状与研究进展[J].工程科学学报, 2019, 41(01): 22-42.

[2] 彭自然,王顺豪,肖伸平,等.基于KAInformer的电动汽车动力电池SOC&SOH估算[J/OL].电工技术学报,2025，1-17.

[3] 闫金定. 锂离子电池发展现状及其前景分析[J]. 航空学报, 2014, 35(10): 2767-2775.

[4] 卢云帆.锂电池荷电状态估计与剩余寿命预测研究[D].安徽理工大学,2023.

[5] 申彩英,左凯.基于开路电压法的磷酸铁锂电池SOC估算研究[J].电源技术,2019,43(11):1789-1791.

[6] 于海芳, 逯仁贵, 朱春波, 等. 基于安时法的镍氢电池SOC估计误差校正[J]. 电工技术学报, 2012, 27(06): 12-18.

[7] 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(06):1995-2010.

[8] 沈佳妮,贺益君,马紫峰.基于模型的锂离子电池SOC及SOH估计方法研究进展[J].化工学报,2018,69(01):309-316.

[9] HOU J, LIU J W, CHEN F W, et al. Robust lithium-ion state-of-charge and battery parameters joint estimation based on an enhanced adaptive unscented Kalman filter, Energy 271 (2023) 126998.

[10] ZHU C Y, WANG S L, YU C M. An improved Cauchy robust correction-sage Husa extended Kalman filtering algorithm for high-precision SOC estimation of Lithium-ion batteries in new energy vehicles, Journal of Energy Storage 88 (2024) 111552.

[11] WU S J, ZAHNG S Z, LI H R, et.al. A totally coupled multi time-scale framework containing full parameters online identification and SOC real-time estimation of lithium-ion battery based on a fractional order model, Journal of Energy Storage, Volume 73, Part C, 2023, 109012.

[12] 夏飞,王志成,郝硕涛,等.基于卡尔曼粒子滤波算法的锂电池SOC估计[J].系统仿真学报,2020,32(01):44-53.

[13] 李晟延, 马鸿雁, 窦嘉铭, 等. 基于PBES-LS-SVM的锂离子电池组SOC预测[J]. 电源技术, 2022, 46(11): 1279-1283.

[14] HU X, LI S E, YANG Y. Advanced machine learning approach for lithium-ion battery state estimation in electric vehicles[J]. IEEE Transactions on Transportation Electrification, 2017, 2(2): 140-149.

[15] 李宁, 何复兴, 马文涛, 等. 基于经验模态分解的门控循环单元神经网络的锂离子电池荷电状态估计[J]. 电工技术学报, 2022, 37(17): 4528-4536.

[16] 汤瑞.基于ELM方法的锂电池充电SOC和SOH状态评估分析[J].现代工业经济和信息化,2024,14(04):126-127 +130.

[17] 赵超,王延峰,林立.基于改进灰狼算法优化核极限学习机的锂电池动力电池荷电状态估计[J].信息与控制,2021, 50(06):731-739.

[18] 陈柏寒,杨威,万文欣,等.基于ISSA-KELM的电动汽车动力电池SOC估算[J].东北电力技术,2024,45(08):1-5+12.

[19] YUAN P J, XU H B, WEI Y Y. SOC Estimation of Lithium-ion Battery Based on Attention Mechanism of EMD-Bi-LSTM Improving by Bayesian Optimization. ACM International Conference Proceeding Series, 2023, 765–774.

[20] HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hubert spectrum for nonlinear and non-stationary time series analysis, Proc. R. Soc. A Math Phys. Eng. Sci. 454 (1998) 903–995,

[21] 邓带雨,李坚,张真源,等.基于EEMD-GRU-MLR的短期电力负荷预测[J].电网技术,2020,44(02):593-602.

[22] 赵凌云,刘友波,沈晓东,等.基于CEEMDAN和改进时间卷积网络的短期风电功率预测模型[J].电力系统保护与控制,2022,50(01):42-50.

[23] 葛亮,袁雪峰,肖小汀,等.基于EMD与小波阈值联合去噪的埋地非金属管道声学定位回波信号提取[J].Journal of Measurement Science and Instrumentation, 2024, 15(04): 417-431.

[24] 孙万麟, 王超. 基于改进的软阈值小波包网络的电力信号消噪[J]. 海军工程大学学报, 2019(4).

[25] 张帅涛,蒋品群,宋树祥,等.基于注意力机制和CNN-LSTM融合模型的锂电池SOC预测[J].电源学报,2024, 22(05):269-277.

[26] ZHANG Z Q, ZHANG L L, LIU D, et al. Measurement and analysis of regional water-energy-food nexus resilience with an improved hybrid kernel extreme learning machine model based on a dung beetle optimization algorithm, Agricultural Systems, Volume 218 (2024).

[27] WANG S, DING J, Qin X B, YAN J. An Improved Whale Optimization Algorithm Based on Nonlinear Convergence Factors and Differential Evolution, 2023 42nd Chinese Control Conference (CCC), Tianjin, China, 2023, 1796-1801.

[28] KALMAN R E. A new approach to linear filtering and prediction problems [J]. Journal of Basic Engineering, 1960, 82(1): 35-45.

[29] LG 18650HG2 Li-ion battery data and example deep neural network xEV SOC estimator script. 2020-3-6.

引用本文

王赛, 沈德智, 丁洁, 程艳云, 基于混合核极限学习机的锂电池状态融合估计[J]. 电气工程与自动化, 2025; 4: (5) : 132-140.

入驻平台

回到顶部

搜索文章: