作者：赵沁峰, 蔡艳平, 王新军

作者单位：火箭军工程大学，陕西 西安 710025

关键词：锂离子电池;等压降放电时间;剩余寿命;鲸鱼优化算法;极限学习机

摘要：

针对标准极限学习机在预测锂离子电池寿命方面算法不稳定以及使用电池容量作为健康因子不易直接测量的问题，提出一种基于等压降放电时间的鲸鱼优化算法（whale optimization algorithm, WOA）优化极限学习机（extreme learning machine, ELM）的锂电池剩余寿命间接预测的方法。首先提取电池等压降放电时间作为锂电池间接健康因子，然后引入鲸鱼优化算法对极限学习机的模型参数进行优化，将电池放电截止电压影响因子融合，建立锂离子电池剩余寿命间接预测模型，最后通过NASA卓越预测中心的锂离子电池数据集B0005、B0006、B0007、B0018对提出的方法进行有效性和稳定性验证。实验结果表明：基于鲸鱼优化算法的极限学习机建立的锂离子电池RUL预测模型与标准极限学习机预测模型相比，操作复杂度较低，多次预测结果稳定，测试精度得到一定提升，模型适用性能较好。

WOA-ELM based indirect prediction of remaining useful life of lithium-ion battery
ZHAO Qinfeng, CAI Yanping, WANG Xinjun
Rocket Force University of Engineering, Xi’an 710025, China
Abstract: Aiming at the problem that the standard extreme learning machine is unstable in predicting the life of lithium-ion batteries and it is not easy to directly measure the battery capacity as a health factor. A method is proposed to optimize the indirect prediction of the remaining life of lithium batteries in ELM based on WOA based on equal voltage drop discharge time. Firstly, the time interval of equal discharging voltage difference is extracted as indirect health factor of lithium battery, and introduce the whale optimization algorithm to optimize the model parameters of the extreme learning machine. The model is established by merging the time interval of equal discharging voltage difference influence factors. Finally, the validity and stability of the proposed method are verified by the lithium-ion battery data set B0005, B0006, B0007, B0018 of NASA’s Prognostics Center of Excellence. The experimental results show that the proposed model has lower operational complexity, stable multiple prediction results, and certain improvement in test accuracy and the applicable performance of the model is good.
Keywords: lithium-ion battery;time interval of equal discharging voltage difference;remaining useful life;whale optimization algorithm;extreme learning machine
2021, 47(9):138-145  收稿日期: 2020-12-15;收到修改稿日期: 2021-02-19
基金项目:
作者简介: 赵沁峰(1997-),男,山西晋城市人,硕士研究生,专业方向为锂离子电池健康状态诊断
参考文献
[1] 陈万, 蔡艳平, 苏延召, 等. 基于改进粒子滤波的锂离子电池剩余寿命预测[J/OL]. 中国测试: 1-6[2021-02-08].https://kns-cnki-net.webvpn.bjmu.edu.cn/kcms/detail/51.1714.TB.20201013.1139.002.html.
[2] ZHANG Q, WHITE R E. Capacity fade analysis of a lithium-ion cell[J]. Journal of Power Sources, 2008, 179(2): 793-798
[3] LIU D T, ZHOU J B, PAN D W, et al. Lithium-ion battery remaining useful life estimation with an optimized Relevance Vector Machine algorithm with incremental learning[J]. Measurement, 2015: 63
[4] 谢文强. 遗传算法优化BP网络的锂电池剩余容量预测[J]. 仪表技术, 2019(1): 35-37+40
[5] 姜媛媛, 刘柱, 罗慧, 等. 锂电池剩余寿命的ELM间接预测方法[J]. 电子测量与仪器学报, 2016, 30(2): 179-185
[6] 刘凯文, 刘聪聪, 李珺凯, 等. 基于GA-ELM模型的锂电池SOH预测[J]. 无线电通信技术, 2019, 45(3): 248-252
[7] 缪家森, 成丽珉, 吕宏水. 基于PSO-ELM的储能锂电池荷电状态估算[J]. 电力工程技术, 2020, 39(1): 165-169+199
[8] 丁阳征, 贾建芳. 改进PSO优化ELM预测锂离子电池剩余寿命[J]. 电子测量与仪器学报, 2019, 33(2): 72-79
[9] HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine: Theory and applications[J]. Neurocomputing, 2006, 70(1–3): 489-501
[10] 李彬, 李贻斌. 基于ELM学习算法的混沌时间序列预测[J]. 天津大学学报, 2011, 44(8): 701-704
[11] 孙弘利, 赵冠都. 基于ELM神经网络的磷酸铁锂电池SOC估算研究[J]. 通信电源技术, 2018, 35(9): 69-71
[12] 宋绍剑, 王志浩, 林小峰. 基于极限学习机的磷酸铁锂电池SOC估算研究[J]. 电源技术, 2018, 42(6): 806-808+881
[13] 毛力, 王运涛, 刘兴阳, 等. 基于改进极限学习机的短期电力负荷预测方法[J]. 电力系统保护与控制, 2012, 40(20): 140-144
[14] MIRJALILI S, LEWIS A. The whale optimization algorithm[J]. Advances in Engineering Software, 2016, 95: 51-67
[15] 律方成, 刘怡, 亓彦珣, 等. 基于改进遗传算法优化极限学习机的短期电力负荷预测[J]. 华北电力大学学报(自然科学版), 2018, 45(6): 1-7
[16] ISHAM M F, LEONG M S, LIM M H, et al. Optimized ELM based on whale optimization algorithm for gearbox diagnosis[J]. MATEC Web of Conferences, 2019, 255
[17] 张吉宣. 锂离子电池剩余寿命预测方法研究[D]. 太原: 中北大学, 2018.