Improved ANN for estimation of power consumption of EV for real-time battery diagnosis

Abstract

In this paper, an artificial neural network (ANN), which estimates the power consumption of an electric vehicle (EV) during the deterioration process of power storage is described. This network provides important information for real-time battery diagnosis, such as state of charge of a Li-Ion battery for an EV or HEV. The data are retrieved from a scaled experiment, based on the JC08 test cycle. The network is presented as a practical alternative to analytical and empirical methods. It can predict the power consumption by an optimal solution and categorize the deterioration of the power storage with high estimation precision and within short time.