Neural Networks applied to Short Term Load Forecasting: A case study

Abstract

A good management of renewable energy systems and energy storage requires Short Term Load Forecasting (STLF). In particular, Artificial Neural Networks (ANN) have proved their ability to cope with data driven nonlinear models. In this paper ANN models are used with input variables such as apartment area, numbers of occupants, electrical appliance consumption and time, in order to achieve a robust model to be used in forecasting energy consumption of general homes. A feed-forward ANN trained with the Levenberg-Marquardt algorithm is tested and their results show a quite accurate model foreseeing that ANNs are a promising tool for STLF.