Predicting Voltage Characteristic of Charging Model for Li-Ion Battery with ANN for Real Time Diagnosis

Abstract

An adaptive characteristic of charging for the rechargeable batteries using the artificial neural network (ANN) method is proposed in this study. This model is based on the voltage charging data of a Li-Ion battery. By the voltage characteristic of charging data that have been used as a parameter to describe the actual quantity of energy, which is a key factor in applications. This estimation is an important and challenging task. The upcoming Electric Vehicle (EV) or Hybrid Electric Vehicle (HEV), are becoming the most important technology in transportation, because of the Eco-friendly and its increasing driving autonomy. The battery performance directly influences the total performance and efficiency of the BMS for this kind of vehicles. As already confirmed the importance of the battery state of charge (SOC) prediction and the nonlinear characteristic between the battery SOC and the external variables, the neural network model is proposed in order to investigate further. In this approach, the ANN can predict the characteristic of the charging model from the batteries, with the optimized model it can be simulated within a short time and with a high accuracy. Which is a different type of approach to the difficult task of SOC of the battery.