Entropy and Correntropy Against Minimum Square Error in Offline and Online Three-Day Ahead Wind Power Forecasting

Author

Bessa, Ricardo J. ; Miranda, Vladimiro ; Gama, João

Author_Institution

INESC Porto, Inst. de Eng. de Sist. e Comput. do Porto, Porto, Portugal

Volume

24

Issue

4

fYear

2009

Firstpage

1657

Lastpage

1666

Abstract

This paper reports new results in adopting entropy concepts to the training of neural networks to perform wind power prediction as a function of wind characteristics (speed and direction) in wind parks connected to a power grid. Renyi´s entropy is combined with a Parzen windows estimation of the error pdf to form the basis of two criteria (minimum entropy and maximum correntropy) under which neural networks are trained. The results are favorably compared in online and offline training with the traditional minimum square error (MSE) criterion. Real case examples for two distinct wind parks are presented.

Keywords

entropy; learning (artificial intelligence); least mean squares methods; load forecasting; power engineering computing; power grids; wind power plants; Parzen window estimation; Renyis entropy; correntropy; minimum square error method; neural network training; offline wind power prediction; online wind power forecasting; power grid; wind characteristics; wind parks; Correntropy; Parzen windows; entropy; neural networks; wind power forecasting;

fLanguage

English

Journal_Title

Power Systems, IEEE Transactions on

Publisher

ieee

ISSN

0885-8950

Type

jour

DOI

10.1109/TPWRS.2009.2030291

Filename

5238550