Wind power forecasting using historical data and artificial neural networks modeling

Authors: K. P. Moustris, D. Zafirakis, K. A. Kavvadias.


One of the main parameters affecting the reliability of the renewable energy sources (RES) system, compared to the local conventional power station, is the ability to forecast the RES availability for a few hours ahead.

To this end, the main objective of this work is the prognosis of the mean, maximum and minimum hourly wind power (WP) 8hours ahead. For this purpose, Artificial Neural Networks (ANN) modeling is applied. For the appropriate training of the developed ANN models hourly meteorological data are used.

These data have been recorded by a meteorological mast in Tilos Island, Greece.

Published in: Power Generation, Transmission, Distribution and Energy Conversion (MedPower 2016), Mediterranean Conference on.
Key words: Wind Power, Forecasting, Artificial Neural Networks
Type: Article
Publisher: IET