Research On Life-cycle SOC Estimation Of Lithium-ion Power Battery Based On AICKF

In recent years,with the increasingly serious environmental pollution and energy crisis,electric vehicles using power batteries as energy sources have ushered in a golden period of development.The accurate estimation of SOC of power battery can improve battery efficiency,extend the driving range of the vehicle,extend battery life,prevent safety problems caused by overcharge and over-discharge effectively.As the battery continues to age,its capacity and internal resistance are also changing,which brings great difficulties to SOC estimation.In this paper,based on the equivalent circuit model,the aging parameters are updated in time through online parameter identification,capacity estimation and other methods,and the AICKF algorithm is used to estimate the SOC.The above methods are verified through experiments.During the battery life cycle,the SOC estimation error can be controlled within 2.5%.The main content of this paper is as follows:(1)Firstly,the battery capacity was calibrated using the battery test system,and the influence of different discharge rates and temperature on the capacity was studied.Then,the OCV-SOC function relationship was fitted based on the results of the OCV experiment,and other battery characteristic experiments such as HPPC experiment and dynamic simulation experiment were carried out for parameter identification and SOC estimation verification.In the battery aging experiment,the capacity degradation under different depths of discharge and rate was studied,and the change of the OCV-SOC relationship under different aging conditions was analyzed.(2)The state space equation of the battery equivalent circuit model was established,and based on the HPPC experiment,the function fitting and the Simulink parameter identification toolbox were used to perform offline parameter identification.Then forgetting factors recursive least squares was used for online parameter identification research,and the identification results were verified under different working conditions.(3)On the basis of the research on the characteristics of Elman neural network,the mind evolution algorithm is used to globally optimize the initial values of the parameters of the Elman neural network to avoid falling into local minimums.Then,the capacity decay characteristics are extracted from the stable charging process,and the gray correlation analysis is used to optimize the voltage interval.The capacity estimation model of the MEA-Elman network is obtained by training with the extracted training data.(4)On the basis of ICKF,the AICKF algorithm is proposed with the absolute value of the terminal voltage estimation error as an adaptive factor to reduce unnecessary iterative corrections,and the process noise and measurement noise of AICKF are corrected through the Sage-Husa estimator.This further improves the accuracy of the algorithm and the efficiency of the algorithm.Then the estimation accuracy and robustness of AICKF are verified under different working conditions.In order to eliminate the influence of battery aging on the accuracy of SOC estimation,the battery model parameters are updated by the online parameter identification and capacity estimation methods introduced in(2)and(3).The simulation results show that the method can effectively eliminate the influence of battery aging and ensure the accuracy of SOC estimation during the battery life cycle.The research results show that compared with CKF,AICKF’s SOC estimation accuracy is significantly improved,and the battery aging parameters are updated in time to ensure the estimation accuracy during the battery life cycle.At the same time,the capacity estimation results provide a reference for the battery health status,which can effectively improve the reliability and safety of the battery.