Research On Fractional Order Theoretical Modeling And State Of Charge Estimation Of Lithium Batteries

At present,in order to alleviate the energy crisis and improve the ecological environment,vigorously developing electric vehicles is the reform trend of the automobile industry under the new situation.For battery management system,battery modeling and accurate estimation of state of charge(SOC)are still the focus and difficulty of technical research,and it is urgent to study new theories and methods to improve them.In this paper,aiming at problem of the accurate modeling and SOC estimation of Li-ion battery,18650 ternary Li-ion battery was selected as the research object.Firstly,the reaction principle and characteristics of Li-ion battery were discussed.Secondly,the second-order fractional order equivalent circuit model of Li-ion battery was established,and the accurate modeling was realized.Finally,the strong tracking fractional-order Kalman filter(STF-FOKF)algorithm was proposed to realize the accurate estimation of battery SOC.The main work includes the following four points:1^st,Li-ion battery was classified and its working principle in charge-discharge state was described.Through simulation and experiment the voltage characteristics,internal resistance characteristics and capacity characteristics of the batteries were further analyzed,which lays the foundation for the modeling of Li-ion battery.2^nd,Owing to accurately reflect the polarization effect and diffusion phenomenon of Li-ion battery during charge-discharge process and reduce the complexity of the model,the second-order fractional order equivalent circuit model was established by analyzing the electrochemical impedance spectrum curve of the battery.Furthermore,through the Hybrid Pulse Power Characterization test,the least square method was used for identifying the component parameters of the established model,and the genetic algorithm was used for identifying the fractional order number of the component;Finally,the superiority of the fractional order model was verified by the pulse discharge test in MATLAB environment,which effectively solves the weighing problem between the complexity and accuracy of the model,and lays the foundation for SOC estimation of Li-ion battery.3^rd,In response to the question that the battery model has uncertainty and the noise statistical characteristics are inaccurate due to the interference of external random factors,the STF-FOKF algorithm which was initialized to estimate battery SOC,based on the established second-order fractional model was proposed.And the equation construction and initialization were carried out based on the second-order fractional model.4^th,So as to verifying the validity and accuracy of STF-FOKF algorithm,a Li-ion battery charge-discharge platform was designed and built to to estimate the SOC of battery.Under the conditions of constant current discharge and NEDC,the proposed STF-FOKF algorithm not only has high accuracy,but also has good convergence and robustness,which meets the research requirements.