Prediction of Lithium-Ion Batteries Output Voltage in Electric Vehicles

Electric vehicles (EVs) are becoming increasingly popular due to their capacity to reduce reliance on oil and reduce greenhouse gas emissions. Among the electric components of an electric vehicle (EV), the battery is considered as the biggest bottleneck. Among the several battery alternatives, lithium-ion batteries are widely used to power electric vehicles. The condition of a battery encompasses several aspects, including the state-of-charge (SoC), state-of-health (SoH), state-of-power (SoP), and state-of-life (SoL). A battery’s state-of-charge (SoC) refers to the proportion of its capacity that is still useable. Changes in the operating circumstances of the electric vehicle (EV) in which the battery is placed have the greatest impact on the SoC. The output voltage of a battery indicates its state of charge. When a battery’s output voltage goes below a set cut-off value, the state-of-charge (SoC) is deemed to be zero. In other words, the SoC is decided by whether the output voltage of the battery exceeds or falls below a certain threshold. In this paper, the neural network (NN) time series analysis is applied to predict the output voltage of a battery is introduced. The proposed approach is base in the past values to predict the actual measurement using Levenberg-Marquardt algorithm to optimize the input values. The objective is to estimate and forecast the output voltage of a battery during the operation of an electric vehicle (EV). The results show that utilizing the NN based Levenberg-Marquardt algorithm for forecasting yields a lower individual absolute error (IAE).