ARIMA vs. Neural networks for wind speed forecasting

Author

Palomares-Salas, J.C. ; De la Rosa, J.J.G. ; Ramiro, J.G. ; Melgar, J. ; Aguera, A. ; Moreno, A.

Author_Institution

Res. Unit PAIDI-TIC-168, Univ. of Cadiz, Cadiz

fYear

2009

fDate

11-13 May 2009

Firstpage

129

Lastpage

133

Abstract

In this paper an ARIMA model is used for time-series forecast involving wind speed measurements. Results are compared with the performance of a back propagation type NNT. Results show that ARIMA model is better than NNT for short time-intervals to forecast (10 minutes, 1 hour, 2 hours and 4 hours). Data was acquired from a unit located in Southern Andalusia (Pentildeaflor, Sevilla), with a soft orography (10 minutes between measurements). This feature is which makes performance of the ARIMA model and the NNT very similar, so a simple forecasting model could be used in order to administrate energy sources. The paper presents the process of model validation, along with a regression analysis, based in real-life data.

Keywords

forecasting theory; neural nets; power engineering computing; regression analysis; time series; velocity measurement; wind power; ARIMA model; model validation; neural networks; regression analysis; time series forecasting; wind speed forecasting; wind speed measurements; Autocorrelation; Load forecasting; Neural networks; Pollution measurement; Predictive models; Weather forecasting; Wind energy; Wind farms; Wind forecasting; Wind speed; ARIMA; Neural networks; Short-term wind speed prediction; time-series; weather forecasting; wind speed;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence for Measurement Systems and Applications, 2009. CIMSA '09. IEEE International Conference on

Conference_Location

Hong Kong

Print_ISBN

978-1-4244-3819-8

Electronic_ISBN

978-1-4244-3820-4

Type

conf

DOI

10.1109/CIMSA.2009.5069932

Filename

5069932