State Estimation Of Lithium Battery Based On Unscented Kalman Filter

Due to the dual challenges of energy crisis and environmental pollution,new energy electric vehicles have become a new direction for the future development of the automobile industry.Compared with the traditional fuel vehicle,the biggest difference of electric vehicle is that its power comes from the power battery,and the research of power battery technology restricts the development level of electric vehicle.The accurate estimation of SOC is an important research content of battery technology.The accurate estimation of SOC can not only extend the battery life,but also more accurately budget the remaining mileage of electric vehicles.However,there are many factors that affect the SOC estimation of battery,and it is very difficult to estimate accurately.It can be seen that the accurate estimation of SOC of battery is not only the key but also the difficulty in the field of electric vehicle development.In this thesis,the research object is lithium iron phosphate battery.In the process of establishing the equivalent circuit model for parameter identification,the consideration of temperature change is added to improve the accuracy of the battery model.In view of the shortcomings of the current Kalman filter method in estimating SOC of battery,an improved unscented Kalman filter algorithm(UKF)is proposed to simplify the calculation process of sigma point selection and improve the SOC of battery Estimation accuracy.The main research contents are as follows:(1)The establishment of equivalent circuit model and parameter identification of lithium battery.This paper introduces the working principle of lithium-ion battery and its internal chemical reaction,and compares several kinds of widely used lithium batteries,and chooses the lithium iron phosphate battery as the research object.The three equivalent circuit models of internal resistance model,PNGV model and Thevenin model are compared.The first-order model of Thevenin is chosen as the equivalent circuit model of this study.The influence of temperature is considered in the model parameter identification.The parameters of the battery are identified by the hybrid pulse power characteristic experiment and the least square method.Finally,the battery model is proved to be very accurate by the simulation experiment Accuracy.(2)SOC estimation of lithium battery based on improved unscented Kalman filter algorithm.Based on the established equivalent circuit model,this paper analyzes the basic principle of Kalman filter algorithm,introduces the commonly used extended Kalman filter(EKF)and UKF Algorithm in battery SOC estimation,improves the unscented Kalman filter algorithm,adds the spherical algorithm when selecting the sigma point,simplifies the calculation process,estimates the SOC of lithium battery,and establishes the algorithm in Matlab environment Through the simulation,the standard UKF algorithm,EKF algorithm and the improved UKF algorithm are compared with the standard value respectively.The results show that the accuracy of the improved UKF algorithm is significantly higher than the other two algorithms.(3)The improved UKF algorithm is used to verify and analyze SOC estimation of power lithium battery under complex practical conditions.The whole vehicle performance simulation platform of pure electric vehicle is built by advisor.Embedding the improved UKF algorithm model into the vehicle model,Then reset some parameters of the battery module,The simulation experiments are carried out in CYC-UDDS cycle and CYC-NEDC cycle.the simulation model based on MATLAB is used to verify the effect of the improved UKF algorithm and EKF algorithm on SOC estimation of battery.The simulation results show that The estimation accuracy of the improved UKF algorithm is higher than that of the EKF algorithm in both cases.At last,the simulation of SOC value of battery module of pure electric vehicle from 100%to 20%in continuous CYC-UDDS cycle is analyzed.The results show that the improved UKF algorithm is effective for SOC estimation of battery in continuous cycle.It can be seen from the experimental results that the improved UKF algorithm still keeps a good accuracy in estimating SOC of battery under relatively complex actual conditions.