Vrla Battery State Of Charge Estimation System Based On GAPSO-LSSVM Algorithm

VRLA batteries are widely used in electric power,communication and other fields.To learn the working status of the battery,to ensure the normal operation of the system,it is necessary to accurately estimate the battery SOC(State of Charge).In order to accurately estimate SOC,this paper did a detailed analysis of the various factors of SOC.Then consider the advantages and disadvantages of the physical modeling method and the system identification method.On this basis,this paper employed the temperature,resistance and the open-circuit voltage as the joint detection value,adopted the LSSVM algorithm to estimate the battery SOC.This method combined the advantages of the resistance and open-circuit voltage method,while the temperature was also among the detection value.Then with the help of the LSSVM algorithm this paper abandoned the unhandy capacity-temperature correction formula.Simulation results show that the proposed algorithm is suitable for the battery SOC estimation.Then,as for parameter optimization problem in LSSVM,this paper used automatic PSO algorithm,successfully getting rid of the impact of subjective factors.Simulation analysis verified the optimization function of PSO.Furthermore,in order to solve the problem that it trended to converge to local optimal solution,the GA was needed to enhance its global search ability.Simulation analysis showed that the proposed GAPSO-LSSVM algorithm is more effective than the traditional one,and the SOC estimation results are excellent.Finally,based on the above detection algorithm,this paper built a hardware platform for it.At the same time,choosing the MATLAB GUI as the development platform,the battery SOC estimation software was built.The entire system is reasonable and functional,which can accurately estimate the battery SOC.