Machine learning techniques for short-term load forecasting

Abstract

Selection of an adequate tool for accurate short-term load forecasting task is becoming more important for electric utilities. Machine learning techniques are proving useful for short-term electricity load forecasting. In this paper we evaluate performance of several machine learning algorithms applied to electricity load datasets. We evaluated performance of SMOreg, and Additive regression algorithms for load forecasting using electricity consumption datasets. We also performed an Artificial Neural Networks (ANN) analysis on short-term load forecasting.