基于白鹭群优化高斯过程回归的锂电池SOH估计方法

doi:10.19799/j.cnki.2095-4239.2025.0021

摘要/Abstract

摘要：

锂离子电池健康状态（State of Health, SOH）估计直接影响锂电池系统的安全性与可靠性，是电池管理系统中的一项重要功能。针对现有数据驱动的SOH估计方法中存在的缺乏不确定性表达、训练数据与测试数据未完全解耦等问题，本文提出了一种基于白鹭群优化（Egret Swarm Optimization Algorithm, ESOA）与高斯过程回归（Gaussian Process Regression, GPR）相结合的SOH估计方法。首先，从同类电池的充电电压、电流及弛豫电压信息中提取与电池老化相关的健康特征，并通过Pearson相关分析法筛选出与电池容量相关性高的健康特征。随后，采用平方指数核函数的高斯过程回归模型进行SOH估计，采用白鹭群优化算法优化GPR模型中超参数。最后，选取同济大学数据集中的NCA和NCM两类电池数据进行实验，验证所提模型的准确性与鲁棒性。实验结果表明，所提方法能够有效提高SOH估计的精度与可靠性。对于所测电池类型中，SOH估计误差的最大RMSE和MAE分别为0.0028和0.22%，相较于传统的GPR模型，误差指标分别降低了58.82%和57.69%。此外，该方法还能够实现SOH区间估计避免高估电池SOH造成的安全问题。

关键词: 锂电池, 健康状态, 白鹭群优化算法, 高斯过程回归, 区间估计

Abstract:

The estimation of the State of Health (SOH) of lithium-ion batteries directly impacts the safety and reliability of battery systems and is a crucial function of battery management systems. Addressing the issues in existing data-driven SOH estimation methods, such as the lack of uncertainty representation and incomplete decoupling of training and testing data, this paper proposes an SOH estimation method based on the Egret Swarm Optimization Algorithm (ESOA) and Gaussian Process Regression (GPR). First, health features related to battery aging are extracted from the charging voltage, current, and relaxation voltage data of similar batteries, and features with high correlation to battery capacity are selected through Pearson correlation analysis. Subsequently, a Gaussian Process Regression model with a squared exponential kernel function is employed for SOH estimation, with the hyperparameters of the GPR model optimized using the Egret Swarm Optimization Algorithm. Finally, the proposed model is validated using NCA and NCM battery datasets from Tongji University to assess its accuracy and robustness. Experimental results demonstrate that the proposed method effectively improves the precision and reliability of SOH estimation. For the tested battery types, the maximum RMSE and MAE of SOH estimation errors are 0.0028 and 0.22%, respectively, representing improvements of 58.82% and 57.69% compared to conventional GPR models. Furthermore, the method enables SOH interval estimation, thereby avoiding safety risks caused by overestimating battery SOH.

Key words: Lithium-ion battery, State of health, Egret swarm optimization algorithm, Gaussian process regression, Interval Estimation

中图分类号:

TM 912

巫春玲, 王立顶, 卢勇, 耿莉敏, 陈昊, 孟锦豪. 基于白鹭群优化高斯过程回归的锂电池SOH估计方法[J]. 储能科学与技术, doi: 10.19799/j.cnki.2095-4239.2025.0021.

Chunling WU, Liding WANG, Yong Lu, Yao MA, Hao Chen, Jinhao Meng. Lithium-ion batteries SOH estimation based on gaussian processed regression optimized by egret swarm optimization[J]. Energy Storage Science and Technology, doi: 10.19799/j.cnki.2095-4239.2025.0021.

图/表 13

表1

图1

图2

图3

图4

表2

图5

图6

图7

表3

图8

图9

表4

参考文献 26

1	杨夯, 黄小庆, 于慎仟, 等. 电化学储能电站主动安全研究[J]. 电力自动化设备, 2023, 43(08):78-87.
	YANG H, HUANG X, YU S, et al. Research on active safety of electrochemical energy storage station[J].Dianli Zidonghua Shebei/Electric Power Automation Equipment, 2023, 43(8):78-87.
2	ZHENG Y, HU J, CHEN J, et al. State of health estimation for lithium battery random charging process based on CNN-GRU method[J]. Energy Reports, 2023, 9:1-10.
3	DINI P, COLICELLI A, SAPONARA S. Review on Modeling and SOC/SOH Estimation of Batteries for Automotive Applications[J]. Batteries-Basel, 2024, 10(1).
4	LI X, ZHANG L, LIU Y, et al. A fast classification method of retired electric vehicle battery modules and their energy storage application in photovoltaic generation[J]. International Journal of Energy Research, 2019, 44(3):2337-2344.
5	赵鹤, 韩策, 程小露, 等. 采用阳极预锂化技术的锂离子电池高倍率老化容量衰减机理研究[J]. 储能科学与技术, 2021, 10(02):454-461.
	ZHAO He, HAN Ce, CHENG Xiaolu, et al. Research on the capacity fading mechanism of high rate aged lithium-ion batteries with anode prelithiation treatment[J]. Energy Storage Science and Technology, 2021, 10(02):454-461.
6	BASIA A, SIMEU-ABAZI Z, GASCARD E, et al. Review on State of Health estimation methodologies for lithium-ion batteries in the context of circular economy[J]. Cirp Journal of Manufacturing Science and Technology, 2021, 32:517-528.
7	CHANG C, WANG Q, JIANG J, et al. Lithium-ion battery state of health estimation using the incremental capacity and wavelet neural networks with genetic algorithm[J]. Journal of Energy Storage, 2021, 38:102570.
8	STROE D-I, SCHALTZ E. Lithium-ion battery state-of-health estimation using the incremental capacity analysis technique[J]. IEEE Transactions on Industry Applications, 2019, 56(1):678-685.
9	CHEN Y, HUANG M. A Method of Battery State of Health Prediction Based on AR-Particle Filter[R]. SAE Technical Paper, 2016.
10	FU J, WU C, WANG J, et al. Lithium-ion battery SOH prediction based on VMD-PE and improved DBO optimized temporal convolutional network model[J]. Journal of Energy Storage, 2024, 87:111392.
11	FENG X, WENG C, HE X, et al. Online State-of-Health Estimation for Li-Ion Battery Using Partial Charging Segment Based on Support Vector Machine[J]. IEEE Transactions on Vehicular Technology, 2019, 68(9):8583-8592.
12	巫春玲,吕晶晶,相里康,等.基于变分模态分解和核极限学习机集成模型的电动汽车锂电池健康状态预测[J/OL].电源学报:1-14,2023-09-21.
	Chunling WU, Jingjing LÜ, XIANGLI Kang, et al. Health State Prediction of Electric Vehicle Lithium Battery based on Integrated Model of Variation Modal Decomposition and Kernel Limit Learning Machine [J/OL]. Journal of Power Supply:1-14, 2023-09-21.
13	Chunling Wu, Juncheng Fu, Xinrong Huang, Xianfeng Xu, Jinhao Meng. Lithium-Ion Battery Health State Prediction Based on VMD and DBO-SVR [J]. Energies, 2023, 16(10), 3993.
14	RAUF H, KHALID M, ARSHAD N. Machine learning in state of health and remaining useful life estimation: Theoretical and technological development in battery degradation modelling[J]. Renewable and Sustainable Energy Reviews, 2022, 156:111903.
15	顾菊平, 蒋凌, 张新松, 等. 基于特征提取的锂离子电池健康状态评估及影响因素分析[J]. 电工技术学报, 2023, 38(19):5330-5342.
	GU J, JIANG L, ZHANG X, et al. Estimation and Influencing Factor Analysis of Lithium-Ion Batteries State of Health Based on Features Extraction[J]. Diangong Jishu Xuebao /Transactions of China Electrotechnical Society, 2023, 38(19):5330-5342.
16	ZHU J, WANG Y, HUANG Y, et al. Data-driven capacity estimation of commercial lithium-ion batteries from voltage relaxation[J]. Nature Communications, 2022, 13(1).
17	李放, 闵永军, 王琛, et al. 基于充电过程的锂电池 SOH 估计和 RUL 预测[J]. 储能科学与技术, 2022, 11(10): 3316-3327.
	LI F, MIN Y J, WANG C, et al. State of health estimation and remaining useful life predication of lithium batteries using charging process[J]. Energy Storage Science and Technology, 2022, 11(10): 3316-3327.
18	张朝龙, 陈阳, 刘梦玲, 等.一种基于ICA-T特征和CNN-LA-BiLSTM 的锂离子电池健康状态估计方法[J]. 储能科学与技术.
	ZHANG ChaoLong, CHEN Yang, LIU MengLing, et al. A state of health estimation method for lithium-ion batteries based on ICA-T
	features and CNN-LA-BiLSTM[J]. Energy Storage Science and Technology.
19	WEN J P, CHEN X, LI X H, et al. SOH prediction of lithium battery based on IC curve feature and BP neural network[J]. Energy, 2022, 261:8.
20	ZHAO Y-P, WANG J-J, LI X-Y, et al. Extended least squares support vector machine with applications to fault diagnosis of aircraft engine[J]. ISA Transactions, 2020, 97:189-201.
21	REEVES B. Noise Modulated GPR: Second Generation Technology[C]. 15th International Conference on Ground Penetrating Radar (GPR), 2014:708-713.
22	李晓燕,李弢,马尽文.高斯过程混合模型在含噪输入预测策略下的煤矿瓦斯浓度柔性预测[J].信号处理,2021,37(11):2031-2040.
	LI Xiaoyan, LI Tao, MA Jinwen. Gaussian process mixture model for flexible prediction of coal mine gas concentration under noisy input prediction strategy[J]. Signal Processing,2021,37(11):2031-2040.
23	徐厚宝, 杨承莲, 张永康. 卡尔曼滤波优化的高斯过程回归模型[J]. 北京理工大学学报:1-8,2024-03-29.
	XU Houbao, YANG Chenglian, ZHANG Yongkang. Optimization Model of Gaussian Process Regression Based on Kalman Filtering. Transactions of BeijingInstitute of Technology:1-8,2024-03-29.
24	SHAMI T M, EL-SALEH A A, ALSWAITTI M, et al. Particle Swarm Optimization: A Comprehensive Survey[J]. IEEE Access, 2022, 10:10031-10061.
25	LI Chaoran,XIAO Fei,FAN Yaxiang, et al. An Approach to Lithium-ion Battery SOH Estimation Based on Convolutional Neural Network[J] Transactions of China Electrotechnical Society,2020,35(19): 4106-4119.
26	Chen Z, Francis A, Li S, et al. Egret swarm optimization algorithm: an evolutionary computation approach for model free optimization[J]. Biomimetics, 2022, 7(4): 144.

电池类型	NCA电池	NCM电池
电池类型	18650	18650
阳极材料	Graphite/Si	Graphite
阴极材料	Li_0.86(Ni_0.86Co_0.11Al_0.03)O₂ (NCA)	Li_0.84(Ni_0.83Co_0.11Mn_0.07)O₂ (NCM)
电解液	-	非水溶剂与LiPF₆
额定电压	-	3.60V
截止电压	2.65V~4.20V	2.50V~4.20V
额定容量	3.5Ah	3.5Ah
电池质量	-	45.0g

电池编号	特征参数
电池编号	HF1 (DCEVI)	HF2 (DCECI)	HF3 (Voltage Max)	HF4 (Voltage Drop)
NCA1	0.9982	0.9985	0.9680	0.9834
NCA2	0.9987	0.9996	0.9803	0.9904
NCA3	0.9968	0.9973	0.9862	0.9880
NCA4	0.9959	0.9990	0.9804	0.9909
NCM1	0.9956	0.9998	0.9932	0.9942
NCM2	0.9962	0.9982	0.9978	0.9978
NCM3	0.9928	0.9980	0.9934	0.9936
NCM4	0.9951	0.9991	0.9942	0.9941

电池编号	计算耗时/s
电池编号	GRU	LSTM	BP	PSO-BP	ESOA-BP	GPR	PSO-GPR	ESOA-GPR
NCA1	0.00999	0.01157	0.17290	0.00984	0.00994	0.01642	0.01437	0.01621
NCA2	0.01382	0.01072	0.10458	0.00705	0.00818	0.01977	0.01682	0.01388
NCA3	0.01272	0.01246	0.01999	0.00950	0.00726	0.00625	0.00639	0.00651
NCA4	0.01597	0.01735	0.01006	0.00804	0.009289	0.01873	0.02061	0.01897

电池编号	计算耗时/s
电池编号	GRU	LSTM	BP	PSO-BP	ESOA-BP	GPR	PSO-GPR	ESOA-GPR
NCM1	0.01371	0.01103	0.00963	0.00920	0.00890	0.00620	0.00810	0.00636
NCM2	0.01471	0.01500	0.00813	0.00684	0.00694	0.01364	0.01161	0.02107
NCM3	0.01046	0.01362	0.00867	0.00761	0.00737	0.01305	0.01221	0.01565
NCM4	0.01847	0.01233	0.00737	000732	0.00690	0.02433	0.02319	0.02668

[1]	陈峥, 彭月, 胡竞元, 申江卫, 肖仁鑫, 夏雪磊. 基于短期充电数据和增强鲸鱼优化算法的锂离子电池容量预测[J]. 储能科学与技术, 2025, 14(1): 319-330.
[2]	周洪, 俞海龙, 王丽平, 黄学杰. 基于BERTopic主题模型的锂电池前沿监测及主题分析研究[J]. 储能科学与技术, 2025, 14(1): 406-416.
[3]	刘迎迎, 张孝远, 刘梦楠, 孙俊章, 张艳. 基于自适应最优组合核函数高斯过程回归的锂电池健康状态区间估计[J]. 储能科学与技术, 2025, 14(1): 346-357.
[4]	郝峻丰, 岑官骏, 乔荣涵, 朱璟, 孙蔷馥, 张新新, 田孟羽, 金周, 詹元杰, 闫勇, 贲留斌, 俞海龙, 刘燕燕, 周洪, 黄学杰. 锂电池百篇论文点评（2024.10.1—2024.11.30）[J]. 储能科学与技术, 2025, 14(1): 388-405.
[5]	陈星光, 沈逸凡, 邵裕新, 郑岳久, 孙涛, 来鑫, 沈凯, 韩雪冰. 面向实车应用的磷酸铁锂电池容量辨识及特异性优化方法研究[J]. 储能科学与技术, 2024, 13(9): 2963-2971.
[6]	陆继忠, 彭思敏, 李晓宇. 基于多特征量分析和LSTM-XGBoost模型的锂离子电池SOH估计方法[J]. 储能科学与技术, 2024, 13(9): 2972-2982.
[7]	陈媛, 章思源, 蔡宇晶, 黄小贺, 刘炎忠. 融合多项式特征扩展与CNN-Transformer模型的锂电池SOH估计[J]. 储能科学与技术, 2024, 13(9): 2995-3005.
[8]	黎耀康, 杨海东, 徐康康, 蓝昭宇, 章润楠. 基于加权UMAP和改进BLS的锂电池温度预测[J]. 储能科学与技术, 2024, 13(9): 3006-3015.
[9]	焦君宇, 张全權, 陈宁波, 王冀钰, 芦秋迪, 丁浩浩, 彭鹏, 宋孝河, 张帆, 郑家新. 电池大数据智能分析平台的研发与应用[J]. 储能科学与技术, 2024, 13(9): 3198-3213.
[10]	刘定宏, 董文楷, 李召阳, 张红烛, 齐昕. 基于RUN-GRU-attention模型的实车动力电池健康状态估计方法[J]. 储能科学与技术, 2024, 13(9): 3042-3058.
[11]	柯学, 洪华伟, 郑鹏, 李智诚, 范培潇, 杨军, 郭宇铮, 蒯春光. 基于多时间尺度建模自动特征提取和通道注意力机制的锂离子电池健康状态估计[J]. 储能科学与技术, 2024, 13(9): 3059-3071.
[12]	刘莹, 孙丙香, 赵鑫泽, 张珺玮. 基于电热耦合模型的宽温域锂离子电池SOC/SOP联合估计[J]. 储能科学与技术, 2024, 13(9): 3030-3041.
[13]	张新新, 岑官骏, 乔荣涵, 朱璟, 郝峻丰, 孙蔷馥, 田孟羽, 金周, 詹元杰, 闫勇, 贲留斌, 俞海龙, 刘燕燕, 周洪, 黄学杰. 锂电池百篇论文点评（2024.06.01—2024.07.31）[J]. 储能科学与技术, 2024, 13(9): 3226-3244.
[14]	黄煜峰, 梁焕超, 许磊. 基于卡尔曼滤波算法优化Transformer模型的锂离子电池健康状态预测方法[J]. 储能科学与技术, 2024, 13(8): 2791-2802.
[15]	杨凯悦, 谢欣兵, 杜晓钟. 基于离散元法的锂电池极片辊压过程探究[J]. 储能科学与技术, 2024, 13(8): 2570-2579.