针对锂电池剩余使用寿命(RUL)难以准确预测的问题,提出一种考虑多种寿命衰退特征与数据时序性的基于粒子滤波改进长短期记忆网络(PF-LSTM)的预测模型,并应用于锂电池的RUL预测。从电池历史充放电老化数据中提取与容量衰退密切相关的健康因子作为LSTM网络的输入,利用PF算法全局优化的能力寻优超参数,包括神经元个数、学习率、节点丢弃率、批尺寸大小、训练步数等6个参数,提高网络的预测能力;引入Dropout层,避免网络过拟合,提高模型的泛化能力。基于NASA PCoE电池数据集进行实验验证,对4块电池在不同预测起始点下的容量估计和寿命情况进行预测,并与经网格搜索的LSTM,SVR等算法进行比较。实验结果表明,PF-LSTM容量估计的RMSE与MAE均在2%以内,且寿命预测误差在3个循环以内,相比于其他算法精度最高。
Abstract
In order to solve the problem that it is difficult to accurately predict the remaining useful life (RUL) of lithium battery, a prediction model of improved long-term and short-term memory network based on particle filter (PF-LSTM) considering various life decay characteristics and data sequence is proposed and applied to the RUL prediction of lithium battery.The health factors closely related to the capacity decline are extracted from the battery historical charge and discharge aging data as the input of the LSTM network, and the global optimization ability of the PF algorithm is used to find the optimal parameters, including the number of neurons, learning rate, node abandonment rate, batch size, training steps and other six parameters to improve the prediction ability of the network; the introduction of Dropout layer to avoid network over-fitting and improve the generalization ability of the model.Based on the experimental verification of NASA PCoE battery data set, the capacity estimation and life of four batteries under different prediction starting points are predicted and compared with LSTM, SVR searched by grid and other algorithms.The experimental results show that the root mean square error RMSE and the average absolute error MAE of PF-LSTM capacity estimation are less than 2%, and the life prediction error is less than 3 cycles, which is the highest compared with other algorithms.
关键词
计量学 /
锂电池 /
剩余使用寿命预测 /
粒子滤波 /
LSTM网络 /
健康因子
Key words
metrology /
lithium battery /
RUL prediction /
particle filtering /
LSTM network /
health factor
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
［1］Xue Q, Shen S Q, Li G, et al. Remaining useful life prediction for lithium-ion batteries based on capacity estimation and Box-Cox transformation ［J］. IEEE Transactions on Vehicular Technology, 2020, 69 (12): 14765-14779.
［2］Dong G Z, Chen Z H, Wei J W, et al. Battery health prognosis using brownian motion modeling and particle filtering ［J］. IEEE Transactions on Industrial Electronics, 2018, 65 (11): 8646-8655.
［3］Zhang Y Z, Xiong R, He H W, et al. Long short-term memory recurrent neural network for remaining useful life prediction of lithium-ion batteries ［J］. IEEE Transactions on Vehicular Technology, 2018, 67 (7): 5695-5705.
［4］Wang Y Z, Ni Y L, Lu S, et al. Remaining useful life prediction of lithium-ion batteries using support vector regression optimized by artificial bee colony ［J］. IEEE Transactions on Vehicular Technology, 2019, 68 (10): 9543-9553.
［5］Ren L, Dong J B, Wang X K, et al. A data-driven auto-CNN-LSTM prediction model for lithium-ion battery remaining useful life ［J］. IEEE Transactions on Industrial Informatics, 2021, 17 (5): 3478-3487.
［6］Xue Z W, Zhang Y, Cheng C, et al. Remaining useful life prediction of lithium-ion batteries with adaptive unscented kalman filter and optimized support vector regression ［J］. Neurocomputing, 2020, 376: 95-102.
［7］Zhang X, Miao Q, Liu Z. Remaining useful life prediction of lithium-ion battery using an improved UPF method based on MCMC ［J］. Microelectronics Reliability, 2017, 75: 288-295.
［8］吴忠强, 胡晓宇, 马博岩, 等. 基于RFF及GWO-PF的锂电池SOC估计 ［J］. 计量学报, 2022, 43 (9): 1200-1207.
Wu Z Q, Hu X Y, Ma B Y, et al. SOC estimation of lithium batteries based on RFF and GWO-PF ［J］. Acta Metrologica Sinica, 2022, 43 (9): 1200-1207.
［9］吴忠强, 王国勇, 谢宗奎, 等. 基于WOA-UKF 算法的锂电池容量与SOC联合估计 ［J］. 计量学报, 2022, 43 (5): 649-656.
Wu Z Q, Wang G Y, Xie Z K, et al. Joint Estimation of the capacity and SOC of lithium battery based on WOA-UKF Algorithm ［J］. Acta Metrologica Sinica, 2022, 43 (5): 649-656.
［10］杨彦茹, 温杰, 史元浩, 等. 基于CEEMDAN和SVR的锂离子电池剩余使用寿命预测 ［J］. 电子测量与仪器学报, 2020, 34 (12): 197-205.
Yang Y R, Wen J, Shi Y H, et al. Prediction of remaining service life of Li-ion battery based on CEEMDAN and SVR ［J］. Journal of Electronic Measurement and Instrumentation, 2020, 34 (12): 197-205.
［11］陈则王, 李福胜, 林娅, 等. 基于GA-ELM的锂离子电池RUL间接预测方法 ［J］. 计量学报, 2020, 41 (6): 735-742.
Chen Z W, Li F S, Lin Y, et al. Indirect prediction method of RUL for lithium ion battery based on GA-ELM ［J］. Acta Metrologica Sinica, 2020, 41 (6): 735-742.
［12］王竹晴, 郭阳明, 徐聪. 基于SAE-VMD的锂离子电池健康因子提取方法 ［J］. 西北工业大学学报, 2020, 38 (4): 814-821.
Wang Z Q, Guo Y M, Xu C. SAE-VMD-based health factor extraction method for lithium-ion battery ［J］. Journal of Northwestern Polytechnical University, 2020, 38 (4): 814-821.
［13］张翾,冯海林. 基于放电过程信息的锂电池剩余寿命预测［J］. 计量学报, 2022, 43 (11): 1492-1500.
Zhang X, Feng H L. Remaining Life Prediction of Lithium Battery Based on Discharge Process Information［J］. Acta Metrologica Sinica, 2022, 43(11): 1492-1500.
［14］Zraibi B, Okar C, Chaoui H, et al. Remaining useful life assessment for lithium-ion batteries using CNN-LSTM-DNN hybrid method ［J］. IEEE Transactions on Vehicular Technology, 2021, 70 (5): 4252-4261.
基金
河北省自然科学基金(F2020203014)
PDF(768 KB)
