Research On State Estimation Of Lithium Battery Based On SREKF

With the continuous development of lithium battery technology,lithium battery has been widely used in real life.However,as the lithium battery management technology is not mature,which greatly limits the development of new energy electric vehicles and related lithium battery industry.The higher the accuracy of state estimation in lithium Battery Management System(BMS),the more guaranteed the safe use,reliable operation and life prediction of lithium batteries will be.State estimation of lithium battery mainly includes State of Charge(SOC)estimation and State of Health(SOH)estimation.This topic studies the SOH and SOC estimation algorithms,and the main work is as follows:Firstly,the second-order RC Equivalent Circuit Model(ECM)was selected as the lithium battery Model,the parameters of ECM were identified under the HPPC condition,and the function relationship between SOC and OCV was determined.In order to solve the problem of divergence and instability of filtering,this paper introduces Square Root(SR)algorithm based on EKF(Extended Kalman Filter)algorithm,and then uses Square Root Extended Kalman Filter(SREKF)algorithm to estimate SOH and SOC.When estimating SOH,the ohmic resistance(R0)representing SOH in ECM is modeled as a state vector,and the state space equation of SOH estimation is established.The SOH of lithium battery can be obtained by estimating the R0 of the battery in real time and utilizing the internal relationship between the lithium battery R0 and SOH.In the estimation of SOC,SOC is directly selected as the state vector,the state space equation of SOC estimation is established,and the SOC of lithium battery is obtained by continuously updating SOC iteratively through SREKF algorithm.Finally,the simulation results under constant current condition and HPPC condition show that compared with EKF algorithm,SREKF algorithm can accurately and reliably estimate SOH and SOC,and it also provides an effective method for SOH estimation of lithium battery.Due to the close relationship between SOH and SOC of lithium battery,Dual Square Root Extended Kalman Filter(Dual-SREKF)algorithm was used to jointly estimate SOH and SOC of lithium battery in order to solve the problem of increasing error caused by constant R0 of SOC estimation.Firstly,according to ECM,the state space equation of SOH and SOC is established Then,one SREKF algorithm is used to estimate R0,and the other SREKF algorithm is used to estimate SOC.The two algorithms are constantly iterating and updating each other.Finally,the verification results under constant current condition and HPPC condition show that compared with SREKF algorithm,the Dual-SREKF algorithm has higher accuracy and stronger stability.