Research On The SOC Estimation Algorithm Based On Lithium Ion Battery

With the rapid development of science and technology today,lithium-ion battery will be regarded as energy source by more and more equipment.The estimation of battery state of charge(SOC,state of charge)plays an increasing important role in lithium-ion battery management system,and its value is also the most basic part.State of charge of battery accurate estimation can not only provide reliable information for the equipment,but also it is helpful for the vehicle management system to control the charging and discharging and energy balance,so it is essential to study the SOC estimation of lithium ion battery.The lithium iron phosphate battery,using current time integral method and Kalman Filter algorithm,for further study on the estimation of state of charge in this paper.The basic working principle to estimate the state of charge of the battery must be clear of lithium iron phosphate,knew the charge and discharge characteristics,and considering the effects of lithium ion battery charged state estimation factors;then according to the basic characteristics of lithium-ion phosphate battery,through the basic model of comparison of various types of batteries,the Thevenin equivalent circuit model was established,and identified the model parameters;after the analysis of several existing battery of SOC estimation methods,this paper proposed an algorithm to estimation the SOC of battery innovation: based on Thevenin equivalent circuit model,joining the efficiency of Kulun which impacts on SOC estimation,combined the improved current time integral method and the open circuit voltage method,using the extended Kalman Filter algorithm(EKF,extended Kalman Filter algorithm)to estimate the lithium iron phosphate battery.In order to verify the advantages and the effectiveness of the improved algorithm,the simulation experiments was carried out by Matlab/Simulink module.The experimental results showed that the proposed algorithm could improve the accuracy of SOC estimation,and thealgorithm could be effectively corrected when there was a large error in the initial state.