Study On SOC Estimation Method Of Power Battery For Electric Vehicle Based On Model

Under the background of energy shortage and environmental pollution,new energy vehicles have become the direction of international automobile industry development.As the power source of new energy vehicles,especially pure electric vehicles,the development of related technologies will restrict the development of new energy vehicles,among which the estimation of state of charge?SOC?has always been the focus of electric vehicle users.Accurate SOC is beneficial to extend the driving course of electric vehicles and the life of the battery pack.This paper is based on the polymer ternary lithium-ion battery with soft package produced by a domestic company as the research object,which mainly have studied the characteristics of ternary battery,the equivalent circuit model,the parameter identification and the algorithm of SOC estimation.The main contents are summarized as follows:?1?This paper has introduced the structure and working principle of ternary battery and designed characterization experiments related to SOC,including different rates of charge and discharge experiment,capacity calibration experiment,open circuit voltage calibration experiment,the efficiency of battery charge and discharge experiment,cycle life experiment,and battery self-discharge characteristics experiments,then amended the definition of SOC based on the influence factors.?2?Based on the comparative analysis of several common battery equivalent circuit models,This paper have selected a second-order charge-discharge RC model that can simulate the external characteristics of the battery accurately.Based on off-line and on-line identification of the equivalent circuit model,a Simulink simulation model based on online identification was built after contrasting two parameter identification methods.Three typical operating conditions were used to verify the accuracy of the online identification results.?3?The principles of KF,EKF,and UKF algorithms are introduced,and the effects of EKF and UKF algorithms for estimating SOC are compared.An improved AUKF algorithm based on Sage-Husa was introduced to adjust the mean and covariance matrix of process noise and measurement noise in real time.This algorithm is combined with the discrete battery equivalent circuit model to estimate the SOC value of the battery under three typical conditions: constant current discharge,constant current and constant voltage charge,and ECE dynamic conditions.The results show that the improved AUKF algorithm can quickly converge to the true value;the steady-state error is also within a small rang;and the robustness of the algorithm is improved.?4?Based on AVL CRUISE,the two main power system modules of motor and battery are selected and the parameters are matched.The complete vehicle model of the pure electric vehicle and the cycle conditions of NEDC₃ and UDDS₃ which are based on NEDC and UDDS cycle conditions are established,and Battery pack voltage,current and other parameters of corresponding conditions are obtained.The parameters were identified and used to estimate the SOC by AUKF algorithm,and compared with the SOC value in AVL CRUISE to verify the feasibility and accuracy of the algorithm to estimate SOC of the electric vehicle battery pack.