SOC Estimation Of Lithium-ion Batteries Based On Adaptive Square Root Volume Kalman Filter Algorithm

In the context of today’s increasingly serious environmental pollution and energy shortage,new energy vehicles have become one of the focal points of the national energy saving and emission reduction and "double carbon" policy.Among them,pure electric vehicles,as an important part of new energy vehicles,have a vital role in the power and safety of the vehicle as the power battery system is the source of power output.In order to ensure the efficient,safe and reliable operation of the power battery system,a real-time battery health management system is needed to monitor,feedback and protect the power battery in real time.The system mainly includes data acquisition,energy management,status estimation,safety protection and communication functions.In the battery health management system,the battery state of charge(SOC)is an important parameter,and its accurate estimation under complex temperature conditions can effectively play the role of BMS.Therefore,in this thesis,a manufacturer’s 18650 Li-ion battery is used as the experimental object,and focuses on the battery modeling and SOC estimation of lithium-ion battery in a wide temperature range:Firstly,to address the problem that it is difficult to estimate the SOC of Li-ion battery in wide temperature range,experiments on the charging and discharging characteristics of Li-ion battery in wide temperature range are carried out,mainly including capacity test,HPPC test,open circuit voltage test and dynamic working condition test,and the different characteristics performance are analyzed according to the experimental data.Secondly,by comparing the advantages and disadvantages of different battery models,the equivalent circuit model is selected.Based on the comprehensive consideration of accuracy and complexity,the order of the equivalent circuit model is determined to be second order,and thus a second-order RC equivalent circuit model with temperature compensation is established.The adaptive least squares with forgetting factor(AFFRLS)method is used to identify the parameters of the battery model,and the simulation is verified in MATLAB based on the experimental data of HPPC with wide temperature range.The results show that the error between the simulation results of the battery model and the actual measurement data is small,which proves that the established battery model is highly accurate and can well reflect the operating characteristics of the battery over a wide temperature range.Then,the SOC estimation algorithm of Li-ion battery based on volumetric Kalman filter in wide temperature range is proposed,and the state equation and measurement equation of the battery model are established based on the discretized battery state space equation.The algorithm program is written in MATLAB environment,and the experimental data collected from DST condition and FUDS condition in wide temperature range are used to simulate and compare the three algorithms of EKF,UKF and CKF.The results show that compared with the other two algorithms,the CKF algorithm has the highest estimation accuracy,the smallest error and the best stability in wide temperature range under two different conditions.Finally,based on the shortcomings of the CKF algorithm,the adaptive square root cubature kalman filter(ASRCKF)is proposed by introducing the square root theory and the idea of covariance matching of the model output residual series.In the MATLAB environment,the convergence of the ASRCKF algorithm is simulated and analyzed based on the constant current discharge condition data.The experimental data obtained from the DST condition and FUDS condition collected in a wide temperature range are used to simulate the estimation accuracy and stability of the ASRCKF algorithm,and both are compared with the CKF algorithm for verification.The results show that the ASRCKF algorithm has better convergence,higher accuracy,and better stability.