Dynamic Vector Model Applied to Wind Speed Prognosis for Eolic Generation
Dynamic characterization of energy availability profiles is paramount for an adequate incorporation of Non-Conventional Renewable Energies. This fact is particularly significant for sizing and design of eolic energy parks. The integration of eolic parks with interconnected systems requires accurate and precise knowledge on maximum and minimum power availability, as well as the moments in which you should expect the aforementioned conditions. Prognosis tools can help to determine the wind speed with a certain degree of reliability, in order to forecast energy availability. In this regard, this article aims at designing and implementing a methodology to generate a dynamic vector-autoregressive-based models for wind speed prognosis. This methodology makes use of techniques such as data clustering, time series statistical analysis and its characterization through time-variant parametric models, for a medium term horizon. The proposed method is able to prognosticate wind speed for a complete day in just one step, instead of classic approaches that repeat several one-step ahead transitions to obtain similar results. The employed methodology facilitates the identification of periodical components of the wind, including daily and seasonal, facilitating the differentiation of data clusters with similar behaviors or tendencies. In order to perform the clustering, seasonal patterns are distinguishable through the use of similar probability distributions. Kullback-Leibler divergence is used as a measure of the difference between the probability distributions, while the K-means algorithm is used for clustering. Finally, for the validation of the design two common methods are implemented: Nielsen Reference Model and an ARMA-GARCH model. Our comparative analysis shows that the proposed method greatly improves the precision and accuracy of the resulting wind forecasting.
How to Cite
K-means Clustering, time series analysis, Kullback-Leibler divergence, Vector auto-regressive model, Wind Forecasting
Kariniotakis, G., Pinson, P., Siebert, N., Giebel, G., & Barthelmie, R. (2004). The state of the art in short-term prediction of wind power-from an offshore perspective. In Proceedings of 2004 SeaTechWeek. October 20-21, Brest, France.
Instituto para la Diversificacion y Ahorro de la Energia (IDAE) (2007). ANEMOS. Estudio sobre la Prediccion Eolica en la Union Europea. Madrid, Spain.
Kullback, S. (1968). Information theory and statistics. Mineola, New York. Dover Publications.
Hershey, J. R., & Olsen, P. A. (2007). Approximating the Kullback Leibler divergence between Gaussian mixture models. IEEE International Conference on Acoustics, Speech and Signal Processing ICASSP 2007, pp. IV-317-IV-320. Honolulu, Unites States of America. doi: 10.1109/ICASSP.2007.366913.
MacQueen, J., (1967). Some methods for classification and analysis of multivariate observations. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics, 281-297, University of California Press, Berkeley, CA, USA. http://projecteuclid.org/euclid.bsmsp/1200512992.
Madsen, H., Pinson, P., Kariniotakis, G., Nielsen, H. A., & Nielsen, T. (2005). Standardizing the performance evaluation of short term wind power prediction models. Wind Engineering, vol. 29, no. 6, pp. 475-489.
Liu, H., Erdem, E., & Shi, J. (2011). Comprehensive evaluation of ARMA–GARCH (-M) approaches for modeling the mean and volatility of wind speed. Applied Energy, 2011, vol. 88, no 3, pp. 724-732.
The Prognostic and Health Management Society advocates open-access to scientific data and uses a Creative Commons license for publishing and distributing any papers. A Creative Commons license does not relinquish the author’s copyright; rather it allows them to share some of their rights with any member of the public under certain conditions whilst enjoying full legal protection. By submitting an article to the International Conference of the Prognostics and Health Management Society, the authors agree to be bound by the associated terms and conditions including the following:
As the author, you retain the copyright to your Work. By submitting your Work, you are granting anybody the right to copy, distribute and transmit your Work and to adapt your Work with proper attribution under the terms of the Creative Commons Attribution 3.0 United States license. You assign rights to the Prognostics and Health Management Society to publish and disseminate your Work through electronic and print media if it is accepted for publication. A license note citing the Creative Commons Attribution 3.0 United States License as shown below needs to be placed in the footnote on the first page of the article.
First Author et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.