Home // COGNITIVE 2014, The Sixth International Conference on Advanced Cognitive Technologies and Applications // View article

Using Reservoir Computing for Forecasting of Wind Power Generated by a Wind Farm

Authors:
Bruna Aguiar
Mêuser Valença

Keywords: Reservoir computing; forecasting of wind power, artificial neural network, MLP

Abstract:
One of the main challenges today is the growing global energy demand. In order to meet this need, the most widely used energy sources are oil, natural gas and coal. The main problem with these sources is due to the fact that, besides being extremely polluting, they are non-renewable sources. Therefore, renewable sources are becoming essential for humanity. Among many of them, the wind is the most promising choice. Wind farms have their potential directly related to the wind power, which requires good estimates of this variable in order to build effective strategies and plans. However, this task presents great difficulties due to the complex characteristics of the wind, such as the high variability of its velocity and direction. This paper aims to use the technique of Reservoir Computing for the prediction of wind power generated by a wind farm and compare its performance with the one produced by the Multi-Layer Perceptron, another type of artificial neural network and the most widely used for this purpose. At the end, it will be possible to analyse the results and conclude which one is more appropriate for predicting wind power.

Pages: 184 to 188

Copyright: Copyright (c) IARIA, 2014

Publication date: May 25, 2014

Published in: conference

ISSN: 2308-4197

ISBN: 978-1-61208-340-7

Location: Venice, Italy

Dates: from May 25, 2014 to May 29, 2014