SOC prediction of Lithium-Ion Battery using Extended Kalman Filter

Abstract

solar energy is become more eminent source of energy among renewable energy sources in order to overcome power shortfall. Energy Storage is also an important subject of renewable power generation, Lithium-Ion (Li-Ion) battery is preferred choice among all batteries due to some advantages, e.g. comparatively better charging and discharging performance, high energy and charge density, and more favorable power support. Battery safety and reliability are ensured usually by battery management system (BMS). Among various parameters in a BMS, State of Charge (SOC) of Li-Ion cells is a key indicator which represents the ratio of the stored energy in the battery to the total energy that the battery can contain. In this paper, state of the art SOC estimation using Ampere-hour (Ah) counting and Extended Kalman filter (EKF) methods have been presented. First, EKF for estimating SOC of Li-Ion battery is mathematically designed. Then electrical battery model is implemented using Ah counting and EKF in MATLAB/Simulink. A comparison of the two methods is given which indicates that the SOC evaluation of the battery using EKF is more accurate than Ah counting method. The error observed from the results of EKF is less than 1%.