Research Of Lithium Battery SOC Algorithm Based On ADVISOR Operation Model

Since beginning of 20th century,people pay more and more attention to the issues of energy consumption and governance of environmental pollution.In this case,the world's major automobile manufacturers are paying high attention to the new energy vehicles,especially the development of pure electric vehicles.Electric vehicles' main energy is supplied by power battery,so the rational usage of the remaining power of power battery is essential for vehicle's performance.And accurate estimate of the battery charge state is one of the important and difficulty in the field of electric vehicle development.For above problems,we will introduce the history of lithium battery and some basic theory at the beginning of this paper,as well as the mature estimation method for battery,SOC.And then we determine to use Central difference method of Kalman filter(Central Differential Kalman filter)to estimate the battery SOC.We access the accuracy of the simulated data through the open circuit voltage characteristic experiment on existing lithium-ion batteries and the Hybrid Pulse Power Characterization(HPPC)experiment,etc.Combining the data fitting method,we provide data support for the Matlab battery model.Furthermore,we state the estimation process of Extended Kalman Filtering Algorithm(EKFA)and the principle and implementation process of Analysis Center Difference Kalman Filtering Algorithm(ACDKFA).By these methods,we obtain the system state,observational equations and derived matrix of Extended Kalman filter Algorithm and Central Difference Kalman Filtering Algorithm.Given the above battery simulation model from experiments,we estimate the SOC of the battery by building Extended Kalman Filter and Central Difference Kalman Filter with Matlab/Simulink.The last part of this paper illustrates the results of lithium-ion batteries status,such as current,under different practical working conditions based on the electric vehicle simulation software using the pure-electric vehicle model built in the electric vehicle simulation software ADVISOR2002,based on Matlab simulation environment.We verify the effectiveness of the estimation of power lithium battery charge state.And the estimation is based on Central Difference Kalman Filter and Extended Kalman Filter designed by Matlab/Simulink.The experimental results show that Central Difference Kalman Filter has obvious advantages in practical battery charge state estimation,comparing to Extented Kalman Filtering Algorithm.