A Comparison of Forecasting Methods for Wind Speed at an Altitude of 120 Meters in Ranong Province

Abstract

The objective of this study was to construct a wind speed forecasting model via the use of 4 statistical methods, namely, Box-Jenkins method, Holt’s exponential smoothing method, Brown’s exponential smoothing method, and damped trend exponential smoothing method. Time series of hourly wind speed at an altitude of 120 meters in Ranong province, which were gathered from the Research Center in Energy and Environment, Thaksin University during September 1, 2015, at time 0.00 o’clock to September 28, 2015, at time 23.00 o’clock of 672 observations were divided into 2 datasets. The first dataset, which consisted of 648 observations from September 1, 2015, at time 0.00 o’clock to September 27, 2015, at time 23.00 o’clock was used for constructing the forecasting models. The second dataset, which consisted of 24 observations from September 28, 2015, at time 0.00 – 23.00 o’clock, was used for comparing the accuracy of the forecasts via the criteria of the lowest mean absolute percentage error (MAPE) and root mean squared error (RMSE). The results indicated that the most accurate method was damped trend exponential smoothing method (MAPE = 11.7318, RMSE = 0.9463).