State of charge estimation of second life batteries using first order Thevenin model

Abstract

Second-life batteries (SLBs) are batteries that have reached the end of their useful life in one application but still have sufficient capacity to be used in another application. This paper aims to provide techniques that can accurately predict the performance of second-life lithium-ion batteries, especially their state of charge (SOC). To achieve the aim of the study, the dynamic behavior of the battery has been investigated through a series of experiments on the SLB. The structure of a battery model for SLB is discussed in detail and a battery model for SLB is presented. This paper uses the first order Thevenin model as the basis of the study. Extended Kalman Filter (EKF) and Particle Filter (PF) is used as an algorithm to estimate the state of charge (SOC) of the battery. The comparison between EKF, PF and coulomb counting method shows that the electrical circuit model has poor performance especially at low state of charge (SOC). The study evaluated the Extended Kalman Filter and Particle Filter algorithms for estimating battery state of charge. The PF algorithm showed better accuracy, with root-mean-square error values of 0.8027% and 1.7040%, indicating its superior performance in predicting for SOC.