A load forecasting hybrid method for an isolated power system

Abstract

This paper presents a load forecasting hybrid model designed for isolated power systems. The proposed model consists of four modules that estimate initially the future load demand and a combination module. Radial basis function neural networks (RBFNNs) are applied to make the initial predictions and multilayer perceptrons (MLPs) are used to combine them. Emphasis is given to the RBFNNs generalization ability developing a self-learning procedure with the Particle Swarm Optimization (PSO) algorithm. Satisfactory results are obtained after the evaluation in the Crete case study.