Neural network and improved method for wind power prediction

DOI

10.2991/icaise.2013.42How to use a DOI?

Keywords

Neural network, genetic algorithm, multi-population genetic algorithm, over- fitted , wind power prediction

Abstract

A single population genetic algorithm is introduced to optimize the weight value and threshold value of the BP network (SGABP) to overcome the issues of the slowly convergence speed and easy to fall into local optimum of BP neural network. Taking the premature of genetic algorithm (GA) into account, a multi-population genetic algorithm is constructed to optimize BP network prediction model, the MPGABP model improves the SGABP model by adding immigrants operator and artificial selection operator; Besides, we improve the generalization ability of system in the machine learning methods, through using the noise sequence method to avoid over-fitted. By using the above models to wind power prediction, we draw a conclusion that the combination of the artificial intelligence algorithms is more effective than a single prediction method to improve the prediction accuracy.

Copyright

© 2013, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Rui Li
PY  - 2013/08
DA  - 2013/08
TI  - Neural network and improved method for wind power prediction
BT  - Proceedings of the 2013 The International Conference on Artificial Intelligence and Software Engineering (ICAISE 2013)
PB  - Atlantis Press
SP  - 199
EP  - 203
SN  - 1951-6851
UR  - https://doi.org/10.2991/icaise.2013.42
DO  - 10.2991/icaise.2013.42
ID  - Li2013/08
ER  -