Modeling And SOC Estimation Of LiFePO₄ Battery Considering Temperature And Hysteresis Characteristics

In recent years,problems such as resource shortage and environmental pollution have become more prominent.Under the background of sustainable development,the development of new energy vehicles is gradually becoming a trend in the automotive industry.Lithium batteries are one of the important power sources of new energy vehicles.Under the environment of vigorous development of new energy vehicles,there are more and more researches on their various characteristics.The State of Charge(SOC)of a power lithium battery is a physical quantity used to describe how much power the battery has left,and it is directly linked to the mileage of the car.In addition,SOC plays an important role in the selection of the charging method of the power lithium battery and the protection of the safety of the power lithium battery.The SOC of the power battery is affected by the accuracy of the battery model and the estimation algorithm.Therefore,this article takes the vehicle’s single lithium iron phosphate battery as the object,and conducts research from two aspects of battery model and SOC estimation algorithm,in order to achieve the purpose of more accurate estimation of battery SOC.Firstly,this paper analyzes the working principle,advantages and disadvantages of existing battery models,and finally chooses Thevenin model with simple parameter identification and clear physical meaning for research.Because the parameters of Thevenin model are fixed and the hysteresis characteristics of Open Circuit Voltage(OCV)are not considered,its application range is limited,and its accuracy is also affected to a certain extent.In this paper,aiming at the changing law of various parameters of power lithium battery in the working process,a Thevenin model in which model parameters change dynamically with the change of environmental temperature and battery SOC is established.In addition,charge-discharge experiments are conducted under different initial conditions of SOC to study the hysteresis characteristics of power lithium batteries.According to the characteristics of the hysteresis characteristics,a mathematical recursive formula method is proposed to describe the hysteresis characteristics of the OCV to establish Variable parameter Thevenin model considering hysteresis.In addition,the accuracy of the filtering algorithm also has an important impact on the estimation accuracy of the power lithium battery SOC.This article is based on the extended EKF(Extended Kalman Filter),and proposes a MVEKF(Modified Covariance Extended Kalman Filter).This method uses the posterior estimated value of SOC to replace the predicted value to calculate the Kalman gain and covariance matrix again,and proceed to the next iteration based on this.Compared with EKF,MVEKF has higher accuracy and robustness.In this paper,a 12A·H square lithium iron phosphate battery is selected for experimental research.Under the conditions of HYNN charging and discharging cabinet and its supporting host computer software,and Partner constant temperature and humidity box,the charging and discharging experiments are carried out under different temperature and SOC conditions.Experimental and simulation results show that the established variable parameter Thevenin model considering hysteresis characteristics has high accuracy and can better reflect the battery status and performance.Compared with EKF,the proposed MVEKF has smaller SOC estimation error and faster convergence speed.