针对实际电池工作在随机局部充放电时传统电池健康状态（SOH）估计算法所需特征无法提取的问题,提出一种基于充电状态下起始电压分段的估计方法,该方法通过分析电池的容量增量曲线提取与起始充电电压点相对应的特征,基于随机森林与相关指标确定最优的特征与特征数量,并以此作为电池初始充电电压分段的依据,在不同的分段区间内采用对应的特征进行SOH估计。随后,使用迁移学习方法解决实际在运行电池数据量较少而无法支撑建模的问题,通过由动态时间规划改进的样本迁移方法TrAdaBoost.R2对电池状态进行估计,结果表明所提方法具有较高的准确性和可靠性。
Abstract
Aiming at the problem that traditional state of health （SOH） estimation algorithms cannot extract the required features during random partial charging and discharging of actual batteries, a segmented estimation method based on the initial voltage under charging state is proposed. This method extracts the features corresponding to the initial charging voltage point by analyzing the capacity increment curve of the battery, determines the optimal features and number of features based on random forest and related indicators, and uses them as the basis for segmenting the initial charging voltage of the battery. Corresponding features are used for SOH estimation in different segmentation intervals. Subsequently, transfer learning methods were used to solve the problem of insufficient data to support modeling in the actual operation of lithium batteries, using the sample transfer method TrAdaBoost.R2 improved by dynamic time warping, and the results show that the proposed method has high accuracy and reliability
关键词
锂离子电池 /
电池健康状态 /
迁移学习 /
随机局部充放电 /
动态时间规整
Key words
lithium-ion battery /
state of health /
transfer learning /
random partial charge and discharge /
dynamic time warping
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 杜燕, 陶骁, 苏建徽, 等. 考虑实际退役电池常用SOC范围的SOH预测[J]. 太阳能学报, 2025, 46(2): 99-105.
Du Y, Tao X, Su J H, et al.SOH prediction considering range of SOC commonly used in actual decommissioned batteries[J]. Acta Energiae Solaris Sinica, 2025, 46(2): 99-105.
[2] 武骥, 方雷超, 刘兴涛, 等. 基于特征优化和随机森林算法的锂离子电池SOH估计[J]. 机械工程学报, 2024, 60(12): 335-343.
Wu J, Fang L C, Liu X T, et al.State of health estimation of lithium-ion battery based on feature optimization and random forest algorithm[J]. Journal of Mechanical Engineering, 2024, 60(12): 335-343.
[3] Gou B, Xu Y, Feng X.State-of-health estimation and remaining-useful-life prediction for lithium-ion battery using a hybrid data-driven method[J]. IEEE Transactions on Vehicular Technology, 2020, 69(10): 10854-10867.
[4] 魏梓轩, 韩晓娟, 李炫. 基于深度神经网络的梯次利用电池健康状态评估[J]. 太阳能学报, 2022, 43(5): 518-524.
Wei Z X, Han X J, Li X.State of health assessment for echelon utilization batteries based on deep neural network[J]. Acta Energiae Solaris Sinica, 2022, 43(5): 518-524.
[5] 熊庆, 邸振国, 汲胜昌. 锂离子电池健康状态估计及寿命预测研究进展综述[J]. 高电压技术, 2024, 50(3): 1182-1195.
Xiong Q, Di Z G, Ji S C.Review on health state estimation and life prediction of lithium-ion batteries[J]. High Voltage Engineering, 2024, 50(3): 1182-1195.
[6] 魏中宝, 阮浩凯, 何洪文. 基于充电数据的多阶段锂离子电池健康状态估计[J]. 北京理工大学学报, 2022, 42(11): 1184-1190.
Wei Z B, Ruan H K, He H W.Multi-stage state of health estimation based on charging phase for lithium-ion battery[J]. Transactions of Beijing Institute of Technology, 2022, 42(11): 1184-1190.
[7] 黄凯,丁恒, 郭永芳, 等. 基于数据预处理和长短期记忆神经网络的锂离子电池寿命预测[J]. 电工技术学报, 2022, 37(15): 3753-3766.
Huang K, Ding H, Guo Y F, et al.Prediction of Remaining Useful Life of Lithium-Ion Battery Based on Adaptive Data Preprocessing and Long Short-Term Memory Network. Transactions of China Electrotechnical Society, 2022, 37(15): 3753-3766.
[8] 施永, 黄宁, 谢缔, 等. 基于自注意力机制和CNN融合的燃料电池故障诊断技术[J]. 太阳能学报, 2025, 46(5): 53-61.
Shi Y, Huang N, Xie D, et al.Fuel cell fault diagnosis technique based on self-attention mechanism and CNN fusion[J]. Acta Energiae Solaris Sinica, 2025, 46(5): 53-61.
[9] Pan S J, Yang Q.A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345-1359.
[10] 杨靖. 基于动态时间弯曲的时间序列相似性搜索技术的研究[D]. 哈尔滨: 哈尔滨工业大学, 2013.
Yang J.Research of similarity search based on dynamic time warping in time series data mining[D]. Harbin: Harbin Institute of Technology, 2013.
[11] Meng Z, Agyeman K A, Wang X Y.Multi-segment state of health estimation of lithium-ion batteries considering short partial charging[J]. IEEE Transactions on Energy Conversion, 2023, 38(3): 1913-1923.
[12] Rakthanmanon T, Campana B, Mueen A, et al.Searching and mining trillions of time series subsequences under dynamic time warping[C]//Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2012: 262-270.
[13] 梁凤勤. 基于迁移学习的锂离子电池健康状态估计方法研究[D]. 成都: 电子科技大学, 2021.
Liang F Q.Research on the method of estimating the health state of lithium-ion battery based on transfer learning[D]. Chengdu: University of Electronic Science and Technology of China, 2021.
[14] Pardoe D, Stone P.Boosting for regression transfer[C]//Proceedings of the 27th International Conference on International Conference on Machine Learning. 2010: 863-870.
[15] Dai W Y, Yang Q, Xue G R, et al.Boosting for transfer learning[C]//Proceedings of the 24th International Conference on Machine Learning. 2007: 193-200.
[16] Rettinger A, Zinkevich M, Bowling M.Boosting expert ensembles for rapid concept recall[C]//Proceedings of the 21st National Conference on Artificial Intelligence-Volume 1.2006: 464-469.
基金
国家电网有限公司科技项目（SGAHDK00CNJS2500178）
PDF(1460 KB)
