Efficient prediction of short term load using Chebyshev functional link artificial neural network

Abstract

In this paper, we have presented an efficient single layer ANN structure called functional link ANN (FLANN) for the prediction of short time load demand. In contrast to a feed forward ANN structure, i.e. a multilayer perceptron (MLP), the FLANN is basically a single layer structure in which non-linearity is introduced by enhancing the input pattern with nonlinear function expansion. In FLANN structure the need of a hidden layer is eliminated and it requires much less computation than that of MLP. In this paper three different functional expansions have been applied. From the simulation studies, it is clear that with the Chebyshev functional expansion, the FLANN network performs best among other competing networks for this challenging and interesting problem. Performance comparisons of the all the networks were carried out through extensive computer simulations.