Research On State Estimation Of Power Batteries Considering Temperature-Hysteresis Effect

Battery state-of-charge(SOC)and state-of-power(SOP)are two of the most significant factors in battery management system,the endurance mileage and instantaneous dynamic performance of electric vehicles can be illustrated by them.To guarantee the safe and stable operation of electric vehicle under complex operating conditions,the accurate and reliable state estimation and optimal management of power batteries are needed.Because of the strong nonlinearity of the battery system,its internal parameters are changed in real time with the working conditions,environmental temperature and other factors,making the battery modeling and state estimation based on this become difficult.Aiming at these problems,a temperaturehysteresis equivalent circuit model for lithium iron phosphate battery is established,and the real-time estimation algorithms of SOC and SOP for batteries are studied in depth.The main research contents are as follows:(1)According to the operating mechanism of lithium iron phosphate battery,the battery characteristics were tested at different temperatures,and the effects of temperature variation and hysteresis on battery modeling were analyzed.Based on the existing battery models,a second-order RC equivalent circuit model based on temperature-hysteresis effect is proposed.According to the variation characteristics of open-circuit voltage and internal resistance at different temperatures,the battery parameters are identified by direct identification method.Finally,the simulation results in Matlab show that the error of the temperature model with hysteresis is smaller and more stable than the one without hysteresis.(2)Due to the time-varying parameters of battery models and the linear error of extended Kalman filter(EKF),the estimation accuracy of SOC is low.Aimed at this problem,the state space equation of SOC estimation is established based on temperature-hysteresis model,and the SOC of batteries is estimated online by using the adaptive Unscented Kalman filter(AUKF)algorithm.The simulation results show that AUKF algorithm has higher accuracy than EKF filtering algorithm,and the error range is stable at constant and variable temperature conditions,which verifies the accuracy of the method.(3)Single restrictive condition results the power battery SOP estimation inaccurate in the charging and discharging process.Aiming at this problem,by analyzing the hybrid pulse power characteristic method(HPPC)and the pulse process under single constraint condition,the combined current-voltage-SOC constraint condition for the battery peak power prediction is proposed.Combining the battery model,the peak current of charge and discharge under multiconstraints is obtained,further the peak power is estimated.The simulation result shows the peak power estimation under the multi-constraint conditions is more reliable compared with the HPPC method.