Research On SOC Estimation And Fast Charging Strategy Of Lithium Power Battery

Power battery as the energy source of electric vehicles,battery management system is mainly responsible for management and monitoring of power batteries.The research object of this paper is lithium-ion battery.Lithium power battery is widely used as a power battery in electric vehicles because of its long life and high specific energy.The accurate estimation of the state of charge(SOC)of the power battery is the basis of the battery management system.The efficient charging strategy is the key to improve the charging efficiency of the battery.Improving the charging efficiency of the lithium ion battery can effectively reduce the energy consumption and charging time.In this paper,the equivalent circuit model construction,SOC estimation and fast charging strategy of lithium ion single cells are studied.The main work of this paper is as follows:Firstly,the research significance and background of SOC estimation of Lithium power battery are expounded.The definition and common estimation method of SOC are introduced.The SOC estimation and fast charging method of Lithium power battery at home and abroad are introduced and analyzed comprehensively.Then,through the battery test platform,the open circuit voltage identification experiment is completed,and the open circuit voltage and SOC curve of the battery are obtained.At the same time,the HPPC mixed pulse power characteristic experiment is used to complete the parameter identification of the battery model.On this basis,the MATLAB simulation model of the first-order RC equivalent circuit of Lithium power battery was built,and the model was verified under the pulse condition.Secondly,the paper studies the application of neural network method and Kalman filtering method in SOC estimation.In terms of neural network method,aiming at the shortcomings of Elman neural network prediction SOC performance and the poor stability of PSO algorithm,slow convergence rate and easy to fall into local convergence,this paper adopt a SOC estimation method based on improved PSO algorithm to optimize Elman neural network,and then use the data collected by the experimental platform to verify the algorithm.In terms of Kalman filtering method,aiming at the insufficient processing ability of EKF algorithm for time-varying noise and computer rounding error,this paper adopt a method of estimating SOC by Sage-Husa's adaptive extended Kalman filter algorithm,which is improved by UD decomposition,and then uses the custom pulse condition and UDDS working current data to verify the algorithm's Performance.For the subsequent charging strategy research,one of the two SOC estimation methods will be used.In this paper,the two SOC estimation methods adopted in this paper are compared with the constant current and constant voltage charging data collected by the experimental platform.And stability meets the needs of practical applications,but the UD-AEKF algorithm performs a little better in estimating SOC performance.Finally,based on UD-AEKF estimation SOC method,the fast charging strategy of lithium ion battery is studied.Taking the energy consumption and time in the charging process as the evaluation index,considering that the internal parameters of the battery are dynamically changing with the SOC,this paper will establish the time-energy coupling model in the charging process based on the established battery equivalent circuit model.And this model was established in MATLAB.After determining the SOC segmentation mode,the genetic algorithm is used to optimize the charging current of each segment of the SOC to obtain an optimized fast charging strategy.The results show that the method can effectively shorten the battery charging time and reduce the battery energy loss.