Short-term wind speed forecasting based on spectral clustering and optimised echo state networks

Predicting the wind speed at multiple time points over a time span between two and 4 h typically requires a multi-input/multi-output model. This study investigates a wind speed forecasting method based on spectral clustering (SC) and echo state networks (ESNs). A wavelet transformation was used to decompose the wind speed into multiple series to eliminate irregular fluctuation. The decomposed series were modelled separately. For every decomposed wind speed series, principal component analysis was used to reduce the number of variables and thus the redundant information among the input variables. SC was used to select similar samples from the historical data to form training and validation sets. An ESN was used to simultaneously predict multiple outputs, and a genetic algorithm was employed to optimise the ESN parameters and ensure the forecast accuracy and the generalisation of the model. The forecasts of the decomposed series were summed to get the wind speed. Tests based on actual data show that the proposed model can simultaneously forecast wind speeds at multiple time points with high efficiency, and the accuracy of the proposed model is significantly higher than that of the traditional models.

[1]  Aoife Foley,et al.  Current methods and advances in forecasting of wind power generation , 2012 .

[2]  Johan A. K. Suykens,et al.  Sparse kernel spectral clustering models for large-scale data analysis , 2011, Neurocomputing.

[3]  Michael I. Jordan,et al.  On Spectral Clustering: Analysis and an algorithm , 2001, NIPS.

[4]  Yongli Wang,et al.  Echo state network with wavelet in load forecasting , 2012, Kybernetes.

[5]  Desheng Dash Wu,et al.  A soft computing system for day-ahead electricity price forecasting , 2010, Appl. Soft Comput..

[6]  Dipti Srinivasan,et al.  A SOM-based hybrid linear-neural model for short-term load forecasting , 2011, Neurocomputing.

[7]  Harald Haas,et al.  Harnessing Nonlinearity: Predicting Chaotic Systems and Saving Energy in Wireless Communication , 2004, Science.

[8]  Ali Deihimi,et al.  Short-term electric load and temperature forecasting using wavelet echo state networks with neural reconstruction , 2013 .

[9]  Jitendra Malik,et al.  Normalized cuts and image segmentation , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[10]  Zehong Yang,et al.  Short-term stock price prediction based on echo state networks , 2009, Expert Syst. Appl..

[11]  Filip De Turck,et al.  Time series classification for the prediction of dialysis in critically ill patients using echo statenetworks , 2013, Eng. Appl. Artif. Intell..

[12]  R. Kavasseri,et al.  Day-ahead wind speed forecasting using f-ARIMA models , 2009 .

[13]  P. McSharry,et al.  A comparison of univariate methods for forecasting electricity demand up to a day ahead , 2006 .

[14]  Wolf-Gerrit Fruh,et al.  Wind forecasting using Principal Component Analysis , 2014 .

[15]  S. H. Cao,et al.  Study of forecasting solar irradiance using neural networks with preprocessing sample data by wavelet analysis , 2006 .

[16]  Ulaş Güntürkün,et al.  Sequential reconstruction of driving-forces from nonlinear nonstationary dynamics , 2010 .

[17]  Shian-Chang Huang,et al.  Integrating spectral clustering with wavelet based kernel partial least square regressions for financial modeling and forecasting , 2011, Appl. Math. Comput..

[18]  Dongxiao Niu,et al.  Short-term wind speed forecasting using wavelet transform and support vector machines optimized by genetic algorithm , 2014 .

[19]  Joao P. S. Catalao,et al.  Short-term wind power forecasting in Portugal by neural networks and wavelet transform , 2011 .

[20]  Jing Shi,et al.  Bayesian adaptive combination of short-term wind speed forecasts from neural network models , 2011 .