Journal of Solar Energy Engineering · 2010 · 12 citations · 16 references
Natural SciencesMeteorologyForecastingTime Series AnalysisWeather ForecastingLong-term ForecastingSeasonal ForecastingTime Series ForecastingTime Series ModelingTime Series PredictionWind ForecastingSeasonal Pattern DetectionSystem MaintenanceTime Series Forecasting For WindTime Series Trend AnalysisWindWind Forecasting Error AnalysisWind Speed ForecastingWind Speed Prediction
This paper presents a time series analysis of historical observations of wind speed in order to project future wind speed trends. For this study, 52 years of data have been used from seven suitable stations across the UK. Four parsimonious models have been employed, and the data were split into two different segments: the training and the validation data sets. During the fitting process, the optimum parameters for each model were determined in order to minimize the mean square error in the predictions. The results suggest that the seasonal pattern in wind speeds is the most important factor but that there is some monthly autocorrelation in the data, which can improve forecasts. This is confirmed by testing the four models with the model having considered both autocorrelation and seasonality achieving the smallest errors. The approach proposed for forecasting wind speeds a month ahead may be deemed useful to suppliers for purchasing base load in advance and to system operators for power system maintenance scheduling up to a month ahead.
16
On a measure of lack of fit in time series models
Greta M. Ljung, George E. P. Box · Biometrika · 1978 · 6K citations
Forecasting methods and applications
Computers & Operations Research · 1993 · 2.4K citations
Business, Mathematics, Meteorology +10