Capacity fade estimation in electric vehicles Li-ion batteries using artificial neural networks

Abstract

Battery performance degrades as the battery ages. For example, the battery capacity fades away after repeatedly cycling the battery. The degradation rate itself depends on many factors such as the depth-of-discharge (DOD), (dis)charge power, temperature, etc. In this paper, the application of artificial neural network (ANN) in estimating lithium-ion (Li-ion) battery capacity fade in electric vehicles (EVs) is investigated. The focus in this paper is on evaluating the performance of ANN-based techniques in estimating the battery capacity fade in order to: reliably estimate the battery state-of-charge (SOC) using the standard coulomb counting method through the battery life, and accurately predict the battery remaining life. Model derivation and experimental verification are presented in this paper.