SOC Estimation And Charging Control Strategy For Li-ion Battery In Electric Vehicles

Power battery is the main energy source of electric vehicles.The battery management system has the function of monitoring battery energy status.This article analyzes the object for lithium-ion batteries,lithium-ion batteries have a long life,low self-discharge rate,high specific energy,etc.,is widely used in electric vehicles.The accurate estimation of the SOC by the battery management system is the basis for improving the battery charging efficiency.The high-efficiency charging control strategy is the key to improving the charging efficiency.Increasing the charging efficiency of the power battery can reduce the energy consumption and reduce the charging time.In recent years,this research group has conducted in-depth studies on model building,SOC estimation,and balance of charge and discharge of new energy automotive single cells and battery packs.Based on laboratory theory and test platform to carry out the study of this topic.In this paper,based on the improved fuzzy logic algorithm,the parameters of the second-order RC battery model are identified.Based on the identified model parameters,the battery SOC is estimated using extended Kalman filter and unscented Kalman filter,and the prediction accuracy is compared under two typical operating conditions.Based on the estimated SOC,the charge control strategy is studied to improve the battery charging efficiency.The main research content of this paper is as follows:1.Analyze the mechanism of lithium-ion batteries and extract important parameters that affect the battery characteristics.Based on the experimental platform of this research group,the battery was qualitatively analyzed.In order to achieve accurate modeling and optimization of the charge control strategy,the charge characteristics and polarization characteristics of the lithium ion battery were analyzed in detail.2.A battery equivalent circuit model was established,and battery parameters were calculated through battery experiments.Based on the sample battery parameters,a decision tree algorithm was used to improve the fuzzy algorithm rule base.The results of on-line identification of battery parameters are obtained,in which the temperature has a greater influence on the identification parameters,and the SOC has less influence on the parameter identification results.The experimental verification algorithm has higher accuracy on-line identification and shorter computation time.3.Based on the results of the second-order RC model and parameter identification of the battery,the SOC estimation of the battery is extended through Kalman filtering and unscented Kalman filtering.Two MATLAB/Simulink simulation models are set up and the simulation results are calculated.Based on two typical operating conditions,the experimental accuracy of the two algorithms is validated.The results show that the untracked Kalman filter has higher prediction accuracy,less steady-state error,and shorter correction time.4.Based on the unscented Kalman predictor SOC,a battery charge control strategy was studied.Taking the energy consumption and temperature rise during the battery charging process as the evaluation index,a temperature-rise energy consumption coupling model is built to determine the SOC segmentation mode and coefficient.The genetic algorithm is used to optimize the charge current of each SOC,and the optimized charge control strategy is obtained.After the charge strategy is optimized,the control strategy is compared with the constant current charge control strategy.This strategy can reduce the battery energy loss.The research content of this paper can provide guidance for battery model parameters online identification,SOC estimation and charging control strategy.The research results have reference value for improving charging efficiency.