The Study On SOC Caculation Of LiFeSO₄Battery Based On Unscented Kalman Filtering

As an indispensable part of huaman life,cars has come into hundreds and thousands family along to the explosive development of the automobile industry.As the number of cars growing rapidly, the energy crisis and environmental pollution problems prompted governments and auto companies to pay more and more attention to electric cars because of its zero-emission. The key technology of electric vehicles is power battery and battery management system. However,the basic element and important precondition of the battery management system is the calculation of the state of charge.Therefore,the research of state of charge is very important.At the beginning, this paper researched the development of the electric vehicles around the world.Through the research of all kinds of battery,founded that LiFeSO4battery is the ideal choice in the current time.In order to have a better understand of LiFeSO4battery,this paper has described the definition of the soc and analysis the performance of LiFeSO4battery.Then, the influence of the discharge current,the temperature of battery,the time of the cycle to the battery performance has been studied.After that,this paper has researched the normally used battery models such as ah integration method,the open circuit voltage method, the neural network method,the kalman filtering method. By means of analysing the advantage and disadvantage of each method,this paper choose the kalman filtering method.At the the same time,compared the characteristic of several developmental method of kalman filitering method like extended kalman filtering, unscented kalman filtering,and responsive kalman filtering.At last, this paper choose the unscented kalman filtering with responsive match method as the calculate solution of the soc.In addition,analyzing of several battery model such as electrochemistry model, neural network model and equivalent electric circuit model has been done to confirm a battery modle.By constrast,this paper choose the second order RC electric circuit model as the model of LiFePO4battery and build the simulation model in the simulink software.At last, this paper acquired the parameteres of the battery model through hppc experiment and verified the battery model.In the end, the unscented kalman filtering has been applied into the soc calculation by building the calculation model in the simulink software. The simulation result and the experiment result has been compared.The result shows that this calculate solution of the soc can efficiently calculate the soc with high accuracy.