Study On Modeling And SOC Estimation Of The Lithium-ion Battery Of Electric Vehicle

For electric vehicles,battery management system(BMS)can make a real-time and accurate estimation of the battery SOC,which is an important premise of vehicles’ normal and safe driving.Therefore,it is necessary to do a further research on battery SOC estimation algorithm.In this paper,we take lithium-ion battery NCR18650A as the research object,use the neural network and genetic algorithm to build the battery model,then estimate the battery SOC.The following aspects are studied:1.First,this paper introduces the basic working principle of lithium-ion battery,and takes the capacity test,HPPC pulse test and UDDS and NYCC driving cycle test on lithium-ion battery NCR18650A.We collect the experimental data,such as the current and voltage of the battery,and analyze the data and its changing trend.2.We make lithium-ion battery model equivalent to a first-order RC model,and identify its four important parameters by using the genetic algorithm(GA).We evaluate the precision of the battery model by comparing the real open circuit voltages with the estimated one.3.Based on the first-order RC equivalent ciruit model,we use the Extended Kalman Filter(EKF)method to estimate the battery SOC.We evaluate the effectiveness of the SOC estimation algorithm which based on EKF method by analyzing the simulation experiment results.4.We apply the two SOC estimation algorithms which are based on the UDDS and NYCC driving cycles respectively to EUDC,US06,1015 and New York bus driving cycles,and design 8 simulation experiments We evaluate the adaptability of the two SOC estimation algorithms by analyzing the simulation experiment results.Through the study of this paper,we prove that making the lithium-ion battery model equivalent to a first-order RC model whose parameters is identified through the genetic algorithm has good accuracy,and the SOC estimation algorithm based on the model can effectively estimate the battery SOC and it has good adaptability under different driving cycles.