Research On Estimation Of State Of Charge Of Lithium-ion Battery

High-energy Lithium-ion battery has the characteristics of high voltage,high energy density and long cycle life.It has been widely applied in electric vehicles,weapons and communication equipments.It is meaningful to implement the research on the estimation of state of charge(SOC)of batteries,which is a key technology for battery management system(BMS)and has much influence on charging-discharging strategy,consuming efficiency,cycling life and safety performance.However,the batteries,with strong time-varying and nonlinear characteristics,and greatly influenced by working conditions,aging and operating environment,the real-time,accurate estimation of their SOC is challenging.This paper takes Li-Co Lithium-ion battery as the research object and studies on the test of battery characteristics,the establishment of equivalent electric circuit model and the algorithm of SOC estimation.Firstly,based on the high-accuracy testing of electrochemical characteristics of batteries,the battery testing platform has been set up and a systematic test program has been designed and implemented,which provided experimental data for the model establishment and algorithm verification.A correction factor of discharge rate for maximum available capacity is put forward and a degradation model of battery capacity is built during its life cycle,the real-time correction of maximum available capacity of batteries is realized,and the influence on the estimation of SOC caused by discrepancy between rated capacity and actual capacity is overcame.Secondly,based on accurately modeling of batteries,the improved multi-order Thevenin equivalent circuit model is built,and the parameters of model are identified.The hysteresis factor is brought in the battery model to describe the variation characteristics of hysteresis voltage,and the effectiveness of hysteresis factor for describing the hysteresis characteristics is verified by simulation.By comparing the accuracies of different order equivalent circuit model and balancing the accuracy and complexity,the improved second-order Thevenin equivalent circuit model is selected for SOC estimation algorithm.Finally,based on the error accumulation of Ah counting and the limitation of noise of Kalman filter algorithm,a new SOC estimation algorithm of Ah counting with particle filter correction is proposed.The effect of proposed algorithm is tested under different current conditions.Compared with the Ah counting and Kalman filter algorithm,the proposed algorithm has a better performance.