摘要
通过对铅酸蓄电池充电阶段特征与电池容量的分析,提出了一种基于相关向量机(RVM)的铅酸蓄电池在线健康状态估计方法。RVM的核函数选取泛化能力较强的混合核函数,用变概率遗传算法(VPGA)对核函数进行参数优化,建立了RVM回归模型。实验结果表明该方法可以实现对铅酸蓄电池电池健康状态的精准估计与在线监测。
Through the analysis of the characteristics of charging stage and the capacity of lead-acid battery,the state of health(SOH)estimation method based on relevance vector machine(RVM)was proposed.The mixture kernels function was adopted so as to enhance the generalization of the RVM model.The variable probability genetic algorithm(VPGA)was used to optimize the relevant parameters of the kernels function,and then,the RVM regression model was established.The experiment results verify that the proposed method can realize accurate estimation and online monitoring of the SOH of the lead-acid battery.
作者
丁一
刘盛终
王旭东
戚艳
霍现旭
胡志刚
DING Yi;LIU Shengzhong;WANG Xudong;QI Yan;HUO Xianxu;HU Zhigang(Electric Power Research Institute of State Grid Tianjin Electric Power Company,Tianjin 300384,China;State Grid Tianjin Electric Power Company,Tianjin 300010,China;Chengdong Power Supply Branch of State Grid Tianjin Electric Power Company,Tianjin 300250,China)
出处
《电气传动》
2021年第22期56-62,共7页
Electric Drive
基金
国网天津市电力公司科技项目(KJ19-1-14)。
关键词
铅酸蓄电池
健康状态
相关向量机
在线估计
lead-acid battery
state of health(SOH)
relevance vector machine(RVM)
online estimate
