Research On SOC Estimation And Management System Of Lithium Battery

With the rapid development of the economy,the consumption of coal,oil,natural gas and other non-renewable energy is increasing day by day,and the energy crisis is increasingly prominent.Meanwhile,the massive use of fossil fuels has seriously polluted the atmosphere,and automobile exhaust is one of the main sources of pollution.In order to alleviate the current energy and environmental crisis,new energy automobiles emerge at the historic moment.As the power source of new energy vehicles,lithium-ion battery performance directly determines the future development prospect of new energy vehicles.Battery management system(BMS)is the core component of new energy vehicles,control the running state of the whole battery pack,and SOC(state of charge)estimation is one of the key functions of BMS.Accurate SOC estimation can effectively reduce the damage caused by overcharging,over discharge and other situations on the battery,can extend the battery life and range,improve the energy utilization efficiency and the overall performance of vehicles,it is of great significance for the study of SOC estimation and BMS.In this paper,the estimation of charge state and the design of battery management system for lithium iron phosphate batteries are studied.Firstly,the research status at home and abroad is summarized,the influence of various battery performance parameters on SOC estimation and the variation law are analyzed.The advantages and disadvantages of the common equivalent circuit model and algorithm for SOC estimation of lithium battery are compared and analyzed.Considering the model complexity and estimation accuracy,the equivalent model of the second-order RC circuit achieves a good balance between the computational complexity and accuracy.Therefore,the second-order RC model is selected as the equivalent circuit model of the lithium battery in this paper,and its effectiveness and accuracy are verified.For this model,parameters are identified by using Least squares fitting and recursive least square method with forgetting factor(on-line parameter identification)respectively and compares the identification precision of the two.The results show that the on-line parameter identification has higher precision,and the model parameters can be adjusted adaptively according to the current actual state of the battery,so as to improve the accuracy of the model.Based on the principle of online parameter identification,the extended kalman filtering algorithm(EKF)was first selected for SOC estimation.In view of the shortcomings of the extended kalman filtering,the adaptive extended kalman filter(AEKF)is obtained by improving it.AEKF enables the noise covariance to be continuously adjusted according to the current state,and the statistical characteristics of system process noise and measurement noise are corrected in real time,overcoming the problems of error divergence and partial solution.Then EKF and AEKF algorithms were used to estimate SOC of lithium battery by combining online parameter identification.In order to verify the accuracy and convergence of the algorithm,three conditions of SOC initial values of 1,0.9 and 0.8 were selected for simulation under constant current condition and dynamic condition.The results show that the on-line parameter identification combined with AEKF has higher precision and can estimate the battery SOC more accurately.Moreover,under the condition that the initial value of SOC is uncertain,the convergence rate is faster and the dynamic adaptability is better.Finally,the STM32 microcontroller is used as the main control module,and the BQ76940 chip is used as the acquisition module to study and design the battery management system,which can realize the collection of parameters such as battery current,voltage,temperature,with functions such as over-current,short-circuit,overvoltage,and under-voltage protection.