Short Term Electric Load Forecasting Using High Precision Convolutional Neural Network

Abstract

In this research work, a novel methodology for the issue of short-term load forecasting (STLF) procedure using convolutional neural network (CNN) is presented. The forecasting outcomes of the proposed CNN model in the field of STLF is compared with the outcomes of autoregressive integrated moving average (ARIMA) model that are most frequently used in time series forecasting arena. Mean average error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE) have been defined as accuracy evaluation parameters which evaluated the performance of both proposed CNN model and ARIMA model. Results obtained from the developed network appear that the strategy has the ability to obtain higher precision and accuracy in load forecasting.