Research On Adaptive SOC Estimation Method And Equalization Technology Of Batteries In Electric Vehicles

With the deterioration of the environment and energy crisis,countries around the world have been looking for clean,efficient and pollution-free new energy vehicles to replace oil-fueled vehicles.Among them,electric vehicles(EVs)have received widespread concern.The core of the electric vehicle is the power battery and the core of the power battery is the battery management system(BMS),which can monitor the battery status,safety alarm and ensure the safety of the electric vehicle.The main functions of the BMS include battery state of charge(SOC)estimation,battery equalization management and thermal management.Among them,battery SOC estimation and battery equalization system in BMS are of great significance to improve battery life and improve vehicle performance.Therefore,this thesis took lithium-ion battery as the research object and proposed a new algorithm for battery SOC estimation and a new strategy for battery equalization.(1)This thesis discussed the internal structure and working principle of the most common lithium-ion battery,lithium iron phosphate battery(LFP)and designed the experimental verification of the battery characteristics of LFP.The relationships between the charge and discharge rate and battery capacity,ambient temperature and battery capacity,as well as cycle times and battery capacity were all summed up.The relationships between open circuit voltage(OCV)and battery SOC,battery internal resistance and battery SOC were also pointed out.These relationships lay the foundation for the design of the battery equivalent circuit model and the algorithm of battery SOC estimation.(2)Based on the battery characteristics obtained above,this thesis added a first-order RC network to improve the PNGV equivalent circuit model and proposed an improved PNGV equivalent circuit model.The accuracy of this improved model was validated by battery experimental data and MATLAB simulation experiments.(3)According to the feature of the improved PNGV circuit model and the battery characteristics,this thesis added the noise covariance adaptive matching part and proposed an Adaptive Unscented Kalman Filter(AUKF)based on Unscented Kalman Filter(UKF).In theory,the algorithm could suppress possible filter divergence and improve accuracy.In this thesis,the accuracy and robustness of the new algorithm were verified under three different working conditions,the constant current conditions,the working condition of different initial values and BBDST working condition.It has been proved that the estimation error of the AUKF algorithm did not exceed 4%,which was more accurate than UKF.Moreover,it could quickly converge to the theoretical value when the initial value was of great difference,which proved the good robustness of the proposed method.(4)Comparing and analyzing various common battery equalization circuits,this thesis proposed a battery equalization topology circuit based on bidirectional Fly-back transformer,discussed the equalization principle and working mode selection of this circuit.Due to the shortcomings of over-equalization,energy loss and the nonlinear characteristics of the battery,this thesis proposed a battery equalization control strategy based on fuzzy control.In order to verify the reliability of the battery equalization system,the battery experiment and MATLAB simulation experiment were carried out on the system.The experimental results showed that the battery equalization circuit and battery balancing strategy proposed in this thesis could effectively improve the consistency of the battery,which had good practical value.