01808nas|a2200253 i 450000500170000000800410001703500150005810000210007324500390009426000520013330000140018549000270019950001390022650200570036552007820042265000210120465000300122570000360125570000480129170000440133971000470138381000590143085600650148920260819090806.0010817s2000    th  uzm rtt  00|  a1eng d  a.b118322770 aSunida Chaokasem10aWind data modeling and forecasting  aBangkok :bAsian Institute of Technology,c2000  a44 leaves1 aThesis ;vno. IM-00-05  aA thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Advanced Technologies  aThesis (M.Eng.) - Asian Institute of Technology 2000  aIn this study, hourly wind speed and direction at Promthep cape in Phuket, are forecasted and simulated using Box-Jenkins and Backpropagation approaches without the use of other external data. It was found that ARIMA models and seasonal ARIMA models are useful for short-range forecasting but are not very good in data generation. Backpropagation network models are good for forecasting and for generating data, which resemble the observed data in terms of the important statistics (mean, variance and skewness coefficient). For forecasting, Box-Jenkins models can perform slightly better than Backpropagation network models. However, for data generation (simulation), Backpropagation models can preserve the statistics of the observed data much better than Box-Jenkins models. 0aWind forecasting 0aWindsxSpeedxMeasurement1 aHuynh, Ngoc Phien,eChairperson1 aSadananda, Ramakoti,eExamination Committee1 aTien, Hoang Le, eExamination Committee2 aRoyal Thai Government, eScholarship Donor2 aAsian Institute of Technology.tThesis ;vno. IM-00-05  3Full-Textuhttp://203.159.5.9/ait-thesis/detail.php?q=B06718