Study On Cell SOC Estimation Method Based On Dual Time-Scale EKPF

Limited to the voltage and capacity of the battery cell,the electric vehicle is powered by a large number of battery cells.Since the SOC of each cell is always inconsistent,the battery cell lifetime is affected and the overall performance of the battery is affected.At present,other estimation methods have problems such as low estimation accuracy,divergence,and high computational complexity.In this dissertation,the main performance of lithium battery is firstly analyzed,and the characteristics and applicable scope of the main battery equivalent circuit model are studied.Finally,the ESC equivalent circuit model is selected as the research basis of the battery cell,The average battery model and the individual difference model were derived,and the state space equation was established.Secondly,this dissertation analyzes the existing SOC estimation methods and analyzes the factors that influence the battery SOC estimation.Because the state space equation of the average model is nonlinear,the state variable SOC is estimated.This paper focuses on the analysis of Extended Kalman Filter(EKF),Particle Filter(PF),Extended Kalman Particle Filter(EKPF).Finally,the EKPF algorithm is used as an estimation algorithm for solving the SOC.Then this chapter proposes an SOC estimation algorithm based on FFRLS and dual time-scale EKPF.FFRLS is a model parameter identification method to ensure the accuracy of the model during the estimation process.The dual time-scale EKPF algorithm can not only ensure the estimation accuracy,but also reduce the amount of calculation through the dual time scale.Finally,the accuracy of the SOC estimation algorithm based on FFRLS and dual timescale EKPF was verified by experiments.First,the FFRLS parameter identification experiment was performed on the battery average model to obtain the real-time model parameters.The single-cell SOC estimation experiment was performed on the series lithium battery pack.The model can be modified based on the real-time parameter values obtained during the estimation.The results show that the proposed method has higher estimation accuracy than the dual time-scale EKF algorithm and the dual time-scale PF algorithm,and shows that the computational complexity of cell SOC estimation will decrease with the increase of the number of cells.Therefore,this method has a high practical value...