Research On SOC Estimation Of Lithium-ion Batteries Based On Extended Kalman Filter

With the development of science and technology,lithium-ion batteries have been widely used in many areas which were used in small portable devices to electric vehicles because of their advantages such as high single cell voltage,fast charging and discharging rate,high safety and high number of cycles.As a major component of the energy system,the accuracy of the estimating battery’s state of charge(SOC)is directly related to the evaluation of the battery’s operating condition,service life and safety performance.Therefore,SOC estimation of Li-ion batteries has become a hot topic of research for many scholars.The SOC estimation based on the equivalent circuit model represented by the Extended Kalman filtering(EKF)algorithm has become one of the important methods for estimating the power of Li-ion batteries with its advantages of high accuracy and no need to store a large amount of data.However,an accurate equivalent circuit model is an important prerequisite for the EKF algorithm to estimate SOC accurately,and the battery itself is affected by many factors such as electrode material,discharge current,and operating temperature.Therefore,establishing the battery equivalent circuit model and sorting out the quantitative relationship of each factor on the accuracy of SOC estimation becomes an important part of using EKF algorithm estimate SOC.In this paper,lithium cobalt-acid battery is studied and an equivalent circuit model has been established.In addition,the influence of external factors on the battery SOC estimation and quantify the law of battery performance parameter changes are considered and summarized.The traditional equivalent circuit model is improved,and compensating silver is added to EKF algorithm to improve the accuracy of battery power estimation.The main research contents include:1.Analysis of the characteristics of lithium-ion batteries made of different materials.It analyzes the internal structure,process and the main parameters of lithium-ion batteries.Changes of battery performance parameters of lithium cobalt-acid batteries are paid more attention at different temperatures and different discharge rates.In addition,the experimental data of SOC of two materials,lithium nickel-acid and lithium iron phosphate,at different temperatures and discharge rates are also studied to analyze the changes in SOC of batteries of different materials when they are affected by temperature and current.2.Establishment of a second-order RC circuit model based on the AIC criterion.Similar battery model as the basis of SOC algorithm measurement,its accuracy is directly related to the accuracy of power estimation.Ordinary qualitative analysis can not determine the accuracy of a certain model,so compare several common equivalent circuit models and use AIC(red pool information)criterion quantitative analysis to select the second-order RC equivalent circuit as the model of EKF algorithm from the multi-order RC circuit.Also the model parameter identification is carried out through the experimental data.3.SOC estimation with the introduction of temperature and discharge rate compensation.Common SOC estimation algorithms are analyzed and compared,and the EKF algorithm combined with second-order RC circuit is established as the method for power estimation,which is verified by steady-state,dynamic and initial value auto-correction simulation.The changes of temperature and discharge rate on the battery performance parameters are analyzed and studied,and the degree of their effects are quantified and the function equations are summarized.An Equilibrium Potential(EP)correction is introduced to the common second-order RC equivalent circuit to compensate for the temperature-induced changes in the battery voltage.In addition,a temperature compensation factor and a discharge rate compensation factor are added to the EKF algorithm to improve the accuracy of battery SOC estimation by comparison with simulation verification.