Multi - Model Adaptive Kalman Filtering SOC Estimation Considering Battery Temperature And Rate Characteristics

The growing global energy crisis and environmental protection issues have gradually exposed the huge defects of traditional fuel vehicles in terms of energy consumption and pollutant emissions.The automotive industry that wants sustainable development requires that electric vehicles continue to advance and gradually replace fuel vehicles.Electric vehicles use the power battery as the power source for the vehicle and can achieve zero emission of exhaust gas.The quality of the battery will directly affect the vehicle’s performance as well as the car’s driving safety.It is the most important technical problem.In order to ens ure excellent performance of the power battery pack and prolong the service life of the battery pack,it is very important for the battery management system to manage and control the battery assembly.The state of charge(SOC)estimation of the battery is the most basic and most important function of the battery management system.Accurate estimation of the SOC is the most critical core technology of the electric vehicle.The battery is a complex nonlinear system.In view of the problem of SOC estimation error caused by the inability of a single battery model to accurately describe the battery characteristics under different conditions,a multi-model adaptive Kalman filter algorithm is designed to estimate the battery SOC.Two models are established to describe the temperature characteristics and rate characteristics of the battery separately.The two models’ estimation are fused to get the final SOC estimation result.First,in order to obtain the temperature characteristics and rate characteristics of the battery pack,the Hybrid Pulse Power Characterization(HPPC)test at different temperatures and the variable current HPPC test were designed;Then the dynamic Thevenin model considering the battery temperature and battery charge-discharge rate is establishe d separately.;Based on the HPPC test,the battery model parameters were identified and the dynamic parameters of the two ba ttery models were determined.Finally,a multi-model adaptive Kalman filtering SOC estimation method considering battery temperature and rate characteristics was proposed;Establishing a simulation model of SOC estimation algorithm,and applying SOC estimation algorithm to dynamic stress test(DST)conditions,the simulation results show that the multi-model filter algorithm in this paper has the advantage of accuracy under the variable current conditions,and the adaptability of the method in a complex environment was verified.The model considering the battery temperature and the battery charge/discharge rate can describe the performance of the battery on the one hand.The multi-model Kalman filter SOC estimation method fuses the temperature and the dynamic characteristics of the battery inside the battery to better explain the complexity of the battery.The environment’s non-linear characteristics make the SOC estimation maintain high accuracy throughout the charge-discharge interval and complex usage environment,improving the accuracy and robustness of SOC estimation.