|
|
|
| Estimation and Influencing Factor Analysis of Lithium-Ion Batteries State of Health Based on Features Extraction |
| Gu Juping1,2, Jiang Ling1, Zhang Xinsong3, Hua Liang3, Cheng Tianyu3 |
1. School of Information Science and Technology Nantong University Nantong 226019 China;
2. School of Electronic & Information Engineering Suzhou University of Science and Technology Suzhou 215009 China;
3. School of Electrical Engineering Nantong University Nantong 226019 China |
|
|
|
Abstract State of health (SOH) estimation is the key function of battery management systems and an important prerequisite for ensuring the safe and stable operation of electrochemical energy storage systems. There are several major problems with existing estimation methods, such as low accuracy, and high complexity. In addition, the interpretability of the estimation model is often ignored resulting in poor reliability of SOH estimation. Recently, some methods extract health features from battery charging curves for SOH estimation to mitigate these problems but most of them suffered from uninformative features and complex features correlation. To bridge the gap, this paper customizes a method that combines estimation and influencing factor analysis of lithium-ion batteries' state of health based on feature extraction.
Firstly, six health features were extracted from the charging curve of lithium-ion batteries, including two new similarity features. they are proposed to quantify the similarity between the initial cycle charging voltage curve and the current cycle charging voltage curve, namely the dynamic time warping distance and the Wasserstein distance. Secondly, the CatBoost method is applied to estimate battery SOH, it is an efficient ensemble learning framework based on gradient boosting decision tree. Finally, the Shapley additive explanations (SHAP) method is introduced to analyze the impact of health features on the estimation results and the coupling relationship between features. The proposed method can further improve the interpretability of the method on the premise of ensuring the accuracy of SOH estimation.
Validation experiments are conducted using multiple batteries from the University of Maryland battery aging dataset. The results show that the proposed SOH estimation method has high accuracy with the average RMSE and MAE are only 2.20% and 1.16%. Compared with the existing methods, the estimation accuracy is improved by more than 12%, and the speed is increased by more than 40%. In addition, the method can still achieve accurate evaluation for batteries with different initial charging SOC and different capacities, which proves that the proposed method has strong adaptability. Finally, the analysis of the evaluation results shows that the similarity feature can effectively characterize the changing trend of the internal resistance in batteries. The CCCT and CVTCT in the health feature have a positive impact on the battery SOH, and DTW and WAS have a negative impact on the battery SOH, and there is a linear coupling relationship between health features.
The following conclusions can be drawn from the analysis: (1) The similarity features obtained based on dynamic time warping distance and Wasserstein distance have a strong correlation with the battery internal resistance, and can be used as new health features for battery SOH estimation. (2) Multiple battery data verification experiments show that the battery SOH estimation accuracy of the CatBoost method has been improved by more than 12% compared to existing methods, and its generalization ability is stronger. (3) The contribution rate and influence degree of each health feature were quantified by the SHAP method, which revealed the coupling relationship between features and improved the interpretability of the model. The analysis results showed that CCCT, CVTCT, DTW and WAS characteristics were the key factors affecting SOH assessment. Among them, CCCT and CVTCT have a positive impact, while DTW and WAS have a negative impact on the estimation model. In addition, there is a negative correlation between the constant current charging time and the dynamic time warping distance features and a positive correlation with the stable time of dV/dt.
|
|
Received: 10 July 2023
|
|
|
|
|
|
[1] 宋天昊, 李柯江, 韩肖清, 等. 储能系统参与多应用场景的协同运行策略[J]. 电力系统自动化, 2021, 45(19): 43-51.
Song Tianhao, Li Kejiang, Han Xiaoqing, et al.Coordinated operation strategy of energy storage system participating in multiple application scenarios[J]. Automation of Electric Power Systems, 2021, 45(19): 43-51.
[2] 吴珊, 边晓燕, 张菁娴, 等. 面向新型电力系统灵活性提升的国内外辅助服务市场研究综述[J]. 电工技术报, 2023, 38(6): 1662-1677.
Wu Shan, BianXiaoyan, Zhang Jingxian, et al. A review of domestic and foreign ancillary services market for improving flexibility of new power system[J]. Transactions of China Electrotechnical Society, 2023, 38(6): 1662-1677.
[3] 李建林, 武亦文, 王楠, 等. 吉瓦级电化学储能电站研究综述及展望[J]. 电力系统自动化, 2021, 45(19): 2-14.
Li Jianlin, Wu Yiwen, Wang Nan, et al.Review and prospect of gigawatt-level electrochemical energy storage power station[J]. Automation of Electric Power Systems, 2021, 45(19): 2-14.
[4] 骆凡, 黄海宏, 王海欣. 基于电化学阻抗谱的退役动力电池荷电状态和健康状态快速预测[J]. 仪器仪表学报, 2021, 41(9): 172-180.
Luo Fan, Huang Haihong, Wang Haixin.Rapid prediction of the state of charge and state of health of decommissioned power batteries based on electro-chemical impedance spectroscopy[J]. Chinese Journal of Scientific Instrument, 2021, 41(9): 172-180.
[5] 武龙星, 庞辉, 晋佳敏, 等. 基于电化学模型的锂离子电池荷电状态估计方法综述[J]. 电工技术学报, 2022, 37(7): 1703-1725.
Wu Longxing, Pang Hui, JinJiamin, et al. A review of soc estimation methods for lithium-ion batteries based on electrochemical model[J]. Transactions of China Electrotechnical Society, 2022, 37(7): 1703-1725.
[6] Markevich E, Salitra G, Aurbach D.Fluoroethylene carbonate as an important component for the formation of an effective solid electrolyte interphase on anodes and cathodes for advanced Li-ion batteries[J]. ACS Energy Letters, 2017, 2(6): 1337-1345.
[7] Wang Dafang, Zhang Qi, Huang Huanqi, et al.An electrochemical-thermal model of lithium-ion battery and state of health estimation[J]. Journal of Energy Storage, 2022, 47: 103528.
[8] Song Ziyou, Hou Jun, Li Xuefeng, et al.The sequential algorithm for combined state of charge and state of health estimation of lithium-ion battery based on active current injection[J]. Energy, 2020, 193: 116732.
[9] Jiang Zeyu, Li Junhong, Li Lei, et al.Fractional modeling and parameter identification of lithium-ion battery[J].Ionics, 2022, 28(9): 4135-4148.
[10] Guo Xiangwei, Kang Longyun, Yao Yuan, et al.Joint estimation of the electric vehicle power battery state of charge based on the least squares method and the Kalman filter algorithm[J]. Energies, 2016, 9(2): 100.
[11] He Hongwen, Xiong Rui, Guo Hongqiang.Online estimation of model parameters and state-of-charge of LiFePO4 batteries in electric vehicles[J]. Applied Energy, 2012, 89(1): 413-420.
[12] 赵靖英, 胡劲, 张雪辉, 等. 基于锂电池模型和分数阶理论的SOC-SOH联合估计[J]. 电工技术学报, 2023, 38(17): 4551-4563.
Zhao Jingying, Hu Jin, Zhang Xuehui, et al.Joint estimation of the SOC-SOH based on lithium battery model and fractional order theory[J]. Transactions of China Electrotechnical Society, 2023, 38(17): 4551-4563.
[13] Zheng Yuejiu, Ouyang Minggao, Lu Languang, et al.Understanding aging mechanisms in lithium-ion battery packs: from cell capacity loss to pack capacity evolution[J]. Journal of Power Sources, 2015, 278: 287-295.
[14] 王萍, 弓清瑞, 张吉昂, 等. 一种基于数据驱动与经验模型组合的锂电池在线健康状态预测方法[J].电工技术学报, 2021, 36(24): 5201-5212.
Wang Ping, Gong Qin, Zhang Jiang, et al.An online state of health prediction method for lithium batteries based on combination of data-driven and empirical model[J]. Transactions of China Electrotechnical Society, 2021, 36(24): 5201-5212.
[15] You G W, Park S, Oh D.Diagnosis of electric vehicle batteries using recurrent neural networks[J]. IEEE Transactions on Industrial Electronics, 2017, 64(6): 4885-4893.
[16] Fan Yaxiang, Xiao Fei, Li Chaoran, et al.A novel deep learning framework for state of health estimation of lithium-ion battery[J]. Journal of Energy Storage, 2020, 32: 101741.
[17] Li Kaiquan, Wang Yujie, Chen Zonghai.A comparative study of battery state-of-health estimation based on empirical mode decomposition and neural network[J]. Journal of Energy Storage, 2022, 54: 105333.
[18] Wang Yujie, Li Kaiquan, Peng Pei, et al.Health diagnosis for lithium-ion battery by combining partial incremental capacity and deep belief network during insufficient discharge profile[J]. IEEE Transactions on Industrial Electronics, 2023, 70(11): 11242-11250.
[19] 周才杰, 汪玉洁, 李凯铨, 等. 基于灰色关联度分析-长短期记忆神经网络的锂离子电池健康状态估计[J]. 电工技术学报, 2022, 37(23) : 6065-6073.
Zhou Caijie, Wang Yujie, Li Kaiquan, et al.State of health estimation for lithium-ion battery based on gray correlation analysis and long short-term memory neural network[J]. Transactions of China Electrotechnical Society, 2022, 37(23): 6065-6073.
[20] Tian Jinpeng, Xiong Rui, Shen Weixiang.State-of-health estimation based on differential temperature for lithium ion batteries[J]. IEEE Transactions on Power Electronics, 2020, 35(10): 10363-10373.
[21] 孙丙香, 苏晓佳, 马仕昌, 等. 基于低频阻抗谱和健康特征融合的锂离子电池健康状态主动探测方法研究[J]. 电力系统保护与控制, 2022, 50(7): 23-30.
Sun Bingxiang, SuXiaojia, Ma Shichang, et al. An active detection method of Li-ion battery health state based on low-frequency EIS and health feature fusion[J]. Power System Protection and Control, 2022, 50(7): 23-30.
[22] Li Xiaoyu, Wang Zhenpo, Zhang Lei, et al. State-of-health estimation for Li-ion batteries by combing the incremental capacity analysis method with grey relational analysis[J]. Journal of Power Sources, 2019, 410/411: 106-114.
[23] Liu Gengfeng, Zhang Xiangwen, Liu Zhiming.State of health estimation of power batteries based on multi-feature fusion models using stacking algorithm[J]. Energy, 2022, 259: 124851.
[24] 肖迁, 穆云飞, 焦志鹏, 等. 基于改进LightGBM的电动汽车电池剩余使用寿命在线预测[J]. 电工技术学报, 2022, 37(17): 4517-4527.
Xiao Qian, Mu Yunfei, Jiao Zhipeng, et al.Improved LightGBM based remaining useful life prediction of lithium-ion battery under driving conditions[J]. Transactions of China Electrotechnical Society, 2022, 37(17): 4517-4527.
[25] 樊亚翔, 肖飞, 许杰, 等. 基于充电电压片段和核岭回归的锂离子电池SOH估计[J]. 中国电机工程学报, 2021, 41(16): 5661-5669.
Fan Yaxiang, Xiao Fei, Xu Jie, et al.State of health estimation of lithium-ion batteries based on the partial charging voltage segment and kernel ridge regression[J]. Proceedings of the CSEE, 2021, 41(16): 5661-5669.
[26] Han Tian, Peng Qinke, Zhu Zhibo, et al.A pattern representation of stock time series based on DTW[J]. Physica A: Statistical Mechanics and Its Applications, 2020, 550: 124161.
[27] Markus Goldstein.FastLOF: An expectation maximization based local outlier detection algorithm[C]//Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012), Tsukuba, Japan, 2012: 2282-2285.
[28] Huang Can, Li Fangxing, Zhan Lingwei, et al.Data quality issues for synchrophasor applications Part II: problem formulation and potential solutions[J]. Journal of Modern Power Systems and Clean Energy, 2016, 4(3): 353-361.
[29] 顾崇寅, 徐潇源, 王梦圆, 等. 基于CatBoost算法的光伏阵列故障诊断方法[J]. 电力系统自动化, 2023, 47(2): 105-114.
Gu Chongyin, Xu Xiaoyuan, Wang Mengyuan, et al.CatBoost algorithm based fault diagnosis method for photovoltaic arrays[J]. Automation of Electric Power Systems, 2023, 47(2): 105-114.
[30] 郑心仕, 梁寿愚, 苏晓, 等. 基于贝叶斯方法与可解释机器学习的负荷特性分析与预测[J]. 电力系统自动化, 2023, 47(3): 56-68.
Zhang Xinshi, Liang Shouyu, Su Xiao.Load characteristic analysis and forecasting based on Bayesian method and interpretable machine learning[J]. Automation of Electric Power Systems, 2023, 47(13): 56-68.
[31] Williard N, He Wei, Osterman M, et al.Comparative analysis of features for determining state of health in lithium-ion batteries[J]. International Journal of Prognostics and Health Management, 2020, 4(1): 1-7.
[32] Meng Huixing, GengMengyao, Han Te. Long short-term memory network with Bayesian optimization for health prognostics of lithium-ion batteries based on partial incremental capacity analysis[J]. Reliability Engineering & System Safety, 2023, 236: 109288.
[33] Choi Y, Ryu S, Park K, et al.Machine learning-based lithium-ion battery capacity estimation exploiting multi-channel charging profiles[J]. IEEE Access, 2019, 7: 75143-75152.
[34] Luo Kai, Zheng Huiru, Shi Zhicong.A simple feature extraction method for estimating the whole life cycle state of health of lithium-ion batteries using transformer-based neural network[J]. Journal of Power Sources, 2023, 576: 233139.
[35] 周航, 程泽, 弓清瑞, 等. 基于TCN编码的锂离子电池SOH估计方法[J]. 湖南大学学报(自然科学版), 2023, 50(4): 185-192.
Zhou Hang, Cheng Ze, Gong Qingrui, et al.SOH estimation method of lithium-ion battery based on TCN encoding[J]. Journal of Hunan University (Natural Sciences), 2023, 50(4): 185-192.
[36] Zhu Jiangong, Wang Yixiu, Huang Yuan, et al.Data-driven capacity estimation of commercial lithium-ion batteries from voltage relaxation[J]. Nature Communications, 2022, 13(1): 1-10. |
|
|
|
2023, Vol. 38 
