Waste Reduction and Environmental Protection in EV Load Forecasting

Abstract

This essay focuses on the use of Gated Recurrent Unit (GRU) model to forecast electric vehicle (EV) energy demand to bolster the energy grid alone and its sustainability. The system that is proposed to predict the energy demand in a 24-hour time frame is the processing of the historical EV charging data. The model training is conducted using time-series data and evaluated using the metrics of mean squared error (MSE) and mean absolute error (MAE). Results indicate that the GRU model can be used to model the trends of energy demand and help to proactively operate the grid and increase the penetration of renewable energy.