Research And Design Of SOC Estimation Method And Battery Management System Of Lithium-ion Battery Based On Gas-liquid Dynamics Model

The battery management system is one of the core systems in electric vehicles.An efficient and stable battery management system can effectively ensure the performance,safety and life of the power battery.The accuracy of the battery model and SOC estimation will directly affect the performance of the battery management system.Therefore,the research on battery model and SOC estimation has important theoretical significance and practical value.The power battery is a time-varying system with a high degree of non-linearity,and its state changes with the temperature and the number of cycles during use,which gives great challenge to the modelling and management of LIBs.This article has carried out researches on LIBs modelling and management,and the specific contents include:(1)A gas-liquid dynamics battery model is established.The battery is simulated by a gas-liquid system which is modeled by a gas-liquid dynamic formula.Then the original gas-liquid model is studied and analyzed,and an improvement plan is proposed.The battery thermal model is established to obtain the temperature distribution inside the battery,and then the true equivalent average temperature of the battery is obtained and input to the original model.Finally,the genetic algorithm is used to identify the parameters of the final model,and the accurate modeling of the power battery is completed.(2)The recursive least squares algorithm with forgetting factor(FFRLS)is used to estimate and correct the model parameters online to ensure the real-time accuracy of the model parameters.Then based on the Kalman filter theory,combined with the gasliquid dynamics model,the model state equation and observation equation are established,and SOC estimation methods based on extended Kalman filter and cubature Kalman filter are designed respectively.Finally,a joint estimation algorithm based on FFRLS and Kalman filters(KFs)is proposed to ensure the accuracy and robustness of SOC estimation.(3)A test bench is built to carry out basic capacity calibration and multi-condition testing of the power battery.The estimation accuracy and robustness of the battery model and SOC estimation algorithm are analyzed and researched.The results show that the comprehensive performance of the SOC estimation algorithm using FFRLSCKF filter is the most excellent,with a maximum estimation error of 2.34%,which has the best estimation effect and algorithm stability.(4)A master-slave integrated battery management system is designed.According to the development process,the hardware,basic software,and main control logic design of the BMS have been completed successively,and the design of the controller boot program based on the UDS protocol has been completed.The basic functions of the battery management system are verified in accordance with the requirements of regulations,and the results showed that the battery management system meets the requirements of the national standard and has good industrial application value.Then the performance of the proposed algorithm in the battery management system is tested and analyzed.The results show that although the estimation error of the proposed algorithm in BMS is enlarged,it still can keep within 4%,and its calculation cost is relatively low.So,it has better advancement and industrial application value.