Short Term Load Forecasting Using A Neural Network Based Time Series Approach

This paper introduces a new neural network architecture - the periodic nonlinear ARIMA (PNARIMA) model. This is a neural network variation of the linear ARIMA model, which is designed for short term load prediction. We begin the paper by making linear predictions of the electric load using ARIMA models. Then we develop the PNARIMA predictor. Both predictors are tested using load data from Batam, Indonesia. The results show that the PNARIMA predictor is better than the ARIMA predictor for all testing periods. This demonstrates that there are nonlinear characteristics of the load that cannot be captured by ARIMA models. In addition, we demonstrate that a single model can provide accurate predictions throughout the year, demonstrating that load characteristics do not change substantially between the wet and dry seasons of the tropical climate of Batam, Indonesia.