Training multilayer perceptrons with the extended Kalman algorithm
Advances in neural information processing systems 1
Fundamentals of neural networks: architectures, algorithms, and applications
Fundamentals of neural networks: architectures, algorithms, and applications
Kalman Filtering and Neural Networks
Kalman Filtering and Neural Networks
Neural Networks for Pattern Recognition
Neural Networks for Pattern Recognition
Hi-index | 0.00 |
The introduction of a large quantity of wind generators on the Portuguese electric grid, will produce the effect of having a large percentage of installed power whose production is not controllable. The consequence is a major difficulty to the grid operator in dealing with power availability and oscillations in the frequency. There is the urgent need of a reliable tool for estimating the expected value of the daily power produced by the wind generators in order to elaborate hourly and daily forwarding-dispatches [2] and [3]. Artificial neural networks (ANN) are being used as a model able to predict the average hourly wind speed. However most of the work applying neural networks to wind speed prediction uses Multi-Layer Perceptrons (MLP) or the recurrent version of them [4],[5],[6] and [7]. This work introduces Radial Basis Function networks (RBF) for wind speed prediction showing that this model of neural networks are more suitable for the task at hand, in terms of on-line decisions, and more efficient to train than MLP. The experiments are made with real-world data.