[Journal papers] [Conference papers] [PhD Theses] [MSc Theses]

 

Conference paper
Neural-network-based ultra-short-term wind forecasting
EWEA 2014 Annual Event
, 2014
ABSTRACT:
In recent years rapid growth of wind power generation in many countries around the world has highlighted the importance of wind prediction. In this work neural networks are used for ultra-short-term wind prediction. In many instances reported in the literature neural network exhibit poor performance - very often because no complexity reduction methods were considered. To that end, in this paper two input variable selection algorithms based on partial mutual information are compared for further use with nonlinear models such as neural networks. Performance improvements of the proposed prediction system are compared to neural networks without input variable selection, and validated for locations near Split, Croatia. The use of neural network drastically outperforms simple persistence estimator on 3 hour horizon.
BibTeX entry:
@inproceedings \{Dalto2014_535,
author = \{Djalto, M. AND Va\v{s}ak, M. AND Baoti\'{c}, M. AND Matu\v{s}ko, J. AND Horvath, K.},
title = \{Neural-network-based ultra-short-term wind forecasting},
booktitle = {
EWEA 2014 Annual Event
},
year = \{2014}
}

 

 

 

 

 

Home
About Us
People
Visitors
Groups
Projects
Publications
Software
Courses
Laboratory
Seminars
Students
Matlab
Alumni
Links