Deep neural network based load forecast

Wan He

COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(3) 258-262

State Grid Energy Research Institute, State Grid Corporation of China

Accurate electrical load forecast has great economic and social value. In this paper, we study deep neural networks based load forecast approaches. We first analyse the critical features related to load forecast. Then we present details of deep neural networks and pre-training technologies, including RBM pre-training and discriminative pre-training. We compare the performances of different neural network models and show the advantages of the proposed methods using a rather large data set of loads.