基于LSSVM的直流电源系统蓄电池剩余容量在线检测  

Online Detection of Battery Residual Capacity in DC Power Supply System Based on LSSVM

作　　者：刘海峰 万宁坤 苏琨 单保涛 王浩彬 LIU Haifeng;WAN Ningkun;SU Kun;SHAN Baotao;W ANG Haobin(State Grid Hebei Electric Power Co,Ltd.Xiongan New Area Power Supply Company,Xiongan 071800,China;Hebei 518 Intelligent Technology Co.,Ltd.,Handan 056107,China)

机构地区：[1]国网河北省电力有限公司雄安新区供电公司,河北雄安071800 [2]河北五一八智能科技有限公司,河北邯郸056107

出　　处：《河北电力技术》2024年第1期28-33,共6页Hebei Electric Power

基　　金：国网河北省电力有限公司科技项目(kjcb2022-019)。

摘　　要：为了解决现有蓄电池剩余容量检测方法存在检测精度较低、检测耗时较长的问题,提出基于LSSVM的直流电源系统蓄电池剩余容量在线检测方法。分析直流电源系统蓄电池剩余容量影响因素,获取蓄电池电压、内阻、温度等运行数据,由采样录波单元实现数据采集和信号的完全隔离,通过智能检测单元在分析蓄电池容量影响因素后,构建了基于LSSVM的蓄电池剩余容量检测模型,选取开路电压、内阻、温度参数作为基于LSSVM的蓄电池剩余容量检测模型的输入,利用灰狼优化算法求解惩罚系数、误差以及核系数的最优解,输出蓄电池剩余容量检测结果,实现蓄电池剩余容量在线检测。实验结果表明:本方法的检测误差控制在±0.01之内,检测时间保持在2 s以下,具有较高的检测精度和检测效率,具有较好的实际应用性能。In order to solve the problems of low detection accuracy and long detection time in existing battery residual capacity detection methods,an online detection method for battery residual capacity in DC power systems based on LSSVM is proposed.Analyze the factors affecting the remaining capacity of DC power system batteries,obtain operational data such as battery voltage,internal resistance,temperature,etc.The data acquisition and signal complete isolation are achieved by the sampling and recording unit.After analyzing the factors affecting battery capacity through the intelligent detection unit,a battery remaining capacity detection model based on LSSVM is constructed.Open circuit voltage,internal resistance,and temperature parameter are used as the input for the battery residual capacity detection model based on LSSVM.The grey wolf optimization algorithm is used to solve the optimal solution of penalty coefficient,error,and kernel coefficient,output the battery residual capacity detection results,and achieve online detection of battery residual capacity.The experimental results show that the detection error of the method proposed in this paper is controlled within±0.01,and the detection time is kept below 2 seconds.It has high detection accuracy and efficiency,and good performance in practical application.

关 键 词：LSSVM 直流电源系统 蓄电池 剩余容量 灰狼优化算法 信息交互 

分 类 号：TM912[电气工程—电力电子与电力传动]