Title :
One-Month Ahead Prediction of Wind Speed and Output Power Based on EMD and LSSVM
Author :
Xiaolan, Wang ; Hui, Li
Author_Institution :
Sch. of Electr. Eng., Lanzhou Univ. of Technol., Lanzhou, China
Abstract :
Wind speed is a kind of non-stationary time series, it is difficult to construct the model for accurate forecast. The way improving accuracy of the model for predicting wind speed up to one-month ahead has been investigated using measured data recorded by wind farm. A forecasting method based on empirical mode decomposition (EMD) and least square support vector machine (LSSVM) is proposed in this paper. The non-stationary time series is decomposed into several intrinsic mode functions (IMF) and the trend term. The different LSSVM models to forecast each IMF are built up. These forecasting results of each IMF are combined to obtain the final forecasting result. Considering the power characteristics, unit efficiency and the operate condition of the generators, the one-month ahead forecasted output power of the wind power plant can be obtained.
Keywords :
least squares approximations; support vector machines; time series; weather forecasting; wind; wind power; empirical mode decomposition; forecast; intrinsic mode functions; least square support vector machine; nonstationary time series; wind output power prediction; wind speed prediction; Character generation; Least squares methods; Power generation; Predictive models; Support vector machines; Velocity measurement; Wind energy generation; Wind farms; Wind forecasting; Wind speed; empirical mode decomposition(EMD); intrinsic mode function(IFM); least square support vector machine (LSSVM); wind power; wind speed forecasting;
Conference_Titel :
Energy and Environment Technology, 2009. ICEET '09. International Conference on
Conference_Location :
Guilin, Guangxi
Print_ISBN :
978-0-7695-3819-8
DOI :
10.1109/ICEET.2009.571
