Battery state-of-charge estimating using Adaptive Extended Kalman Filter with Fuzzy modelling of the nominal battery capacity

Abstract

The observable battery parameters like terminal voltage, current and temperature couldn’t give an accurate idea about state of charge (SOC) and state of health (SOH), it is why large number of techniques and algorithms have been proposed to predict the internal parameters (internal resistances Rint, capacitance, and open circuit voltage VOC) which are known as SOC and SOH indicators. In this paper we use an adaptive extended Kalman filter (AEKF) to estimate on-line the internal parameter and SOC based on Thevenin equivalent circuit model. In order to identify the real energy available in the battery, the AEKF algorithm is coupled with Fuzzy modelling of the nominal battery capacity (Cn) that depends on the debited battery current. Experience shows that our approach contributes accurately to estimate the SOC.