Naslov
Neural-network-based ultra-short-term wind forecasting

Autori
Đalto, Mladen ; Vašak, Mario ; Baotić, Mato ; Matuško, Jadranko ; Horvath, Kristian

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the European Wind Energy Association 2014 Annual Event (EWEA 2014) / - , 2014, 1-8

Skup
European Wind Energy Association 2014 Annual Event (EWEA 2014)

Mjesto i datum
Barcelona, Španjolska, 10.03.2014. - 13.03.2014

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
neural networks ; ultra short term ; partial mutual information

Sažetak
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.

Izvorni jezik
Engleski

Znanstvena područja
Temeljne tehničke znanosti

POVEZANOST RADA