Lithium Battery SOC Estimation By CKF

Electric vehicles in China have developed rapidly in recent years.Lithium Titanate battery has the advantages of good performance,long service life and high safety.Therefore,it has become an ideal power device for electric vehicles.Accurate SOC can effectively prevent battery overcharge or overdischarge,ensure the safe application of battery,realize the use battery energy by highest efficiency,and provide the data scheme for the subsequent active equalization research.It has great significance to predict the remaining mileage of electric vehicle and maintain battery pack.Because of the current drift current in the sensor,the estimation algorithm and battery model use the data with errors,which leads to the deviation of SOC estimation.The chemical reaction principle and working characteristics of lithium ion battery are introduced,then highlighted the advantages of lithium titanate battery in application,which provides a reliable experimental object for the subsequent simulation.Then the main SOC estimation algorithms at present are explained,and the algorithms used in subsequent SOC estimation are determined through analysis and comparison.In view of the phenomenon that the estimation of SOC results is affected by drift current,the analysis of the generation of drift current shows that the drift current exists for a long time but has been neglected.Some sensors which have different measurement accuracy were selected to verify the influence of drift current on the SOC estimation results through simulation experiments.The experiment shows that the estimation error is proportional to the drift current.Finally,EKF,UKF and CKF are introduced,and analyzed the performance of the three filters.The filtering accuracy of UKF and CKF is higher than EKF,and proved the stability and convergence of CKF algorithm by mathematical theory when it is dealing with nonlinear systems.Then carrying out the classical charging and discharge experiments of lithium titanate battery.The Drift-Ah integral method and electrochemical noise model were used as state equation and observation equation of the filter.The SOC of the battery was estimated by CKF.By comparing SOC estimation error and calculation by this scheme with other methods,The results show that the proposed Drift-Ah integration method and the improved electrochemical noise model can achieve better SOC estimation outcome.