On-line Supercapacitor State-of-Charge Diagnosis in Electric Vehicle Application

Abstract

Abstract Energy storage systems play an important role in sinking and sourcing of power in an electric vehicle and ensuring operational safety. Supercapacitors are a recent addition to the types of energy storage unit that can be used in an electric vehicle as an ESS because of their high power density, fast charging or discharging, and low internal loss. They can be used in parallel with batteries or fuel cells to form a hybrid energy storage system that makes better use of merits of each component and offsets their drawbacks. This paper, present a feedforward artificial neural networks for supercapacitor state-of-charge diagnosis in vehicular applications. The presented method is evaluated experimentally using a supercapacitor Maxwel model subject to dynamic charge and discharge current profiles and change in ambient temperature. The porposed wavelet neural network shows a dramatically increases of accuracy of state-of-charge estimation.