Study On The Estimation Method Of Lithium Battery State Of Charge For Electric Vehicle

Firstly,the paper describes that an accurate estimation of state of charge is very important for electric vehicles,and presents the research state at home and abroad and compares the advantage and disadvantage of various estimation methods.The basic structure and working principle of lithium battery are introduced.This paper compares four models of battery equivalent.One of the four models named Thevenin is not only simple in structure but also accurate enough and it is selected as the research object.After the battery model is set up,HPPC testing on one battery is processed in order to verify parameters of battery model.we fit the verified parameters with the polynomial and get different parameters with state of charge relationship.In order to verify the accuracy of the model.the same one kind of current input condition is sent into the real battery and battery model and their result is compared to see if they match.Then this paper introduces Ah estimation method of state of charge,and describe the kalman filter and extended kalman filter principle.Based on Thevenin model.An model of estimation of state of charge is set up using extend kalman filter.Aiming at the defects of Ah integral method and extended kalman filter algorithm,this paper proposes the dual kalman filter algorithm to estimate the state of charge.Based on the results of extended kalman and Ah integral estimation algorithm of Soc,which are fused with kalman filter algorithm.Finally,the battery is tested under three different current conditions.and their outputs of state of charge are compared with the measured values.It shows that the results of dual kalman filter compared with the other two algorithm are more accurate and faster to track the true value.