In order to improve the accuracy of short-term wind speed prediction,this paper combines the wavelet transformation and time series method and takes random components into consideration for short-term wind speed prediction.The wavelet transformation was used to stratify the wind speed time series.After that,the high-frequency-varied wind speed time series were predicted by time series Autoregressive Moving Average Model (ARMA),and the low-frequency-varied wind speed time series were predicted by time series continuous method.The final prediction results were the combination of the above two components,and they were further refined by a method of random array.The example verification and the comparison with the prediction by time series method show that the prediction accuracy and prediction stability of the method are significantly improved.
Journal of Inner Mongolia University of Technology(Natural Science Edition)
time series ARMA
time series continuous method
wind speed prediction