The Estimation Of State Of Charge Of Lithium-ion Battery Used On Enhanced Self-correcting Model

In recent years,the energy crisis and the serious problem of environmental pollution,make the development of new energy getting more and more people's attention.As a kind of clean energy,lithium-ion battery is a hot research topic in the field of energy.Although the lithium ion battery production technology is constantly improving,but because of its poor temperature and poor enough to overcharge discharge ability,we must design BMS(Battery Management System)for the process of charge and discharge management of batteries.SOC estimation is one of the necessary work,and also the foundation of other BMS control.When sudden change in current state,Lithium-ion battery produces relaxation and hysteresis phenomenon,based on the model of the battery DC resistance.DC filter let the battery model output a stable current value when its current change into 0 or another constant value,then solves hysteresis effect at the same time.And through the experiment of constant current impulse and dynamic cycling condition,the reliability of the battery was verified.Then,we give the SOC estimation method using extended Calman Filter,based on which we deduction finite difference extended Calman filter algorithm and SOC estimation method based on the self-correcting model and FEDKF.By comparing the simulation results with the EKF algorithm in DC,pulse and complex road conditions,which can be seen that the method proposed in this paper has higher precision.Finally,a battery detection system is designed.The system design use distributing design based on DSP as the upper computer and the STM32 as lower computerthe lower computer.The system can realize the battery voltage,battery current and temperature with real-time display,and transmission data to the upper computer DSP finite difference extended Calman filter by CAN,and at last displays estimated data.