基于SVM的锂电池SOC估算  

Estimation of lithium battery SOC based on SVM

作　　者：骆秀江 张兵[1] 黄细霞[1] 侯程[1] 

机构地区：[1]上海海事大学科学研究院,上海201306

出　　处：《电源技术》2016年第2期287-290,共4页Chinese Journal of Power Sources

基　　金：交通运输部应用基础研究项目(2013329810350);国家自然科学基金资助项目(51209134);上海海事学术新人项目(YXR2015019)

摘　　要：以锂电池SOC作为研究对象,将基于VC维和结构最小化理论为基础的支持向量机(SVM)的方法引入到锂电池SOC的估算中。充分利用支持向量机的对锂电池非线性独特的功能,综合考虑锂电池的电压、温度及电流等因素对SOC的影响,提出了支持向量机估算电池SOC的算法,并将其在锂电池充放电实验中验证。结果表明,支持向量机在估算锂电池SOC时,可以获得更高的估算精度,为电池管理系统提供一种实用的SOC估算方案。A method of estimation of the SOC of lithium battery based on SVM was studied. Support vector machine based on the VC dimension and structural risk minimization theory was used to estimate the SOC of lithium battery.Support vector machine which had unique functions to the nonlinear of lithium battery and the lithium battery voltage,temperature and current on the influence of SOC were comprehensively considered. Then Support vector machine algorithm for estimating the battery SOC was proposed. A Charging and discharging test was used to testify its accuracy. The results indicate that the Support vector machine which is used to estimate the SOC of lithium battery which can acquire higher estimation accuracy. Then the accuracy of SOC estimation which provide a practical solution for the battery management system was improved.

关 键 词：支持向量机 锂电池SOC 非线性 电池管理系统 

分 类 号：TM912[电气工程—电力电子与电力传动]