A comparative study between conventional ARMA and Fourier ARMA in modeling and forecasting wind speed data

Abstract

Wind speed is a fundamental atmospheric variable which plays an important role in energy production industry. Wind speed affects weather forecasting, aircraft and maritime operations, and other numerous effects. As a rising tenders of wind energy, it is vital for power efficacies to plan the adaptation of wind power. Henceforth, an accurate measurement of wind speed prediction is ideal for providing an impression on how the behavior and trend of historical wind pattern and future projected pattern could be. Wind speed forecasting is important for the reliable and efficient operation of the wind power system. The ability of Fourier based-ARMA model was tested to forecast the wind speed data of 3 crucial locations; Senai, Bayan Lepas and Subang. A comparison has been made between conventional method and Fourier analysis method. Fourier-ARMA was found to outperform the conventional ARMA model in predicting the wind speed data for 365 days ahead with just a little error.