SciELO - Scientific Electronic Library Online

 
vol.23 número4Estimating greenhouse gas emissions associated with achieving universal access to electricity for all households in South Africa índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados

Articulo

Indicadores

Links relacionados

  • En proceso de indezaciónCitado por Google
  • En proceso de indezaciónSimilares en Google

Compartir


Journal of Energy in Southern Africa

versión On-line ISSN 2413-3051
versión impresa ISSN 1021-447X

Resumen

SELCUK NOGAY, H.; AKINCI, Tahir Cetin  y  EIDUKEVICIUTE, Marija. Application of artificial neural networks for short term wind speed forecasting in Mardin, Turkey. J. energy South. Afr. [online]. 2012, vol.23, n.4, pp.2-7. ISSN 2413-3051.

Artificial neural network models were used for short term wind speed forecasting in the Mardin area, located in the Southeast Anatolia region of Turkey. Using data that was obtained from the State Meteorological Service and that encompassed a ten year period, short term wind speed forecasting for the Mardin area was performed. A number of different ANN models were developed in this study. The model with 60 neurons is the most successful model for short term wind speed forecasting. The mean squared error and approximation values for training of this model were 0.378088 and 0.970490, respectively. The ANN models developed in the study have produced satisfactory results. The most successful among those models constitutes a model that can be used by the Mardin Electric Utility Control Centre.

Palabras clave : artificial neural network; back propagation; forecasting; wind speed.

        · texto en Inglés     · Inglés ( pdf )