One Day Ahead Prediction of Wind Speed Class by Statistical Models

Silvia Nunnari, Luigi Fortuna, Giorgio Guariso

Abstract


This paper deals with the clustering of daily wind speed time series based on two features: the daily average wind speed and the degree of fluctuation. Daily values of the feature pairs are first classified by means of the fuzzy c-means unsupervised clustering algorithm and then results are used to train a supervised MLP neural network classifier. It is shown that associating to a true wind speed time series a time series of classes allows performing some useful statistics. Further, the problem of predicting the class of daily wind speed 1-step ahead is addressed by using both Hidden Markov Models (HMM) and Non-linear Auto-Regressive (NAR) approaches. The performances of the considered class prediction models are finally assessed in terms of confusion matrices, True Positive rate and True Negative rate, also in comparison with the persistent model.


Keywords


wind speed; time series clustering; fcm algorithm; HMM models; NAR models.

Full Text:

PDF

References


A. Costa, A. Crespo, J. Navarro, G. Lizcano, H. Madsend, E. Feitosa, A review on the young history of the wind power short-term prediction, Renewable and Sustainable Energy Reviews 12 (2008) 1725–1744.

G. Giebel, R. Brownsword, G. Kariniotakis, M. Denhard, C. Draxl, The state of the art in short-term prediction of wind power - a literature overview, Project ANEMOS Deliverable Report D1.1, 2003http://www.anemos-plus.eu/ (2011) 1–110.

J. Jung, R. P. Broadwater, Current status and future advances for windspeed and powerforecasting 31 (2013) 762-777.

T. W. Liao, Clustering of time series data - a survey, Pattern Recognition 38 (2005) 1857–1874.

S. Rani, G. Sikka, Recent techniques of clustering of time series data: A survey, International Journal of Computer Applications 52 (2012) 1–9.

Z. Song, X. Geng, A. Kusiak, C. Xu, Mining markov chain transition matrix from wind speed time series data, Expert Systems with Applications 38 (2011) 10229-10239.

A. Troncoso, S. Salcedo-Sanz, C. Casanova-Mateo, J. R. , L. Prieto, Local models-based regression trees for very short-term wind speed prediction, Renewable Energy 81 (2015) 589–598.

T. Soubdhan, R. Emilion, R. Calif, Classification of daily solar radiation distributions using a mixture of dirichlet distributions, Solar Energy 83 (2009) 1056–1063. doi:10.1016/j.chaos.2008.07.020.

S. I. Colak, M. Demirtas, M. Yesilbudak, A data mining approach: Analyzing wind speed and insolation period data in Turkey for installations of wind and solar power plants, Energy Conversion and Management 65 (2013) 185-197.

L. Fortuna, G. Guariso, S. Nunnari, One day ahead prediction of wind speed class, Proc. of the 2015 International Conference on Renewable Energy Research and Applications (ICRERA), Palermo (Italy), Nov. 2015, pp 965 - 970. DOI:10.1109/ICRERA.2015.7418553.

L. Fortuna, S. Nunnari, A. Gallo, A Typical Day Based Approach To Detrend Solar Radiation, Proc. of the MAED14, November 7, 2014, Orlando, Florida, USA, http://dx.doi.org/10.1145/2661821.2661823.

R. Weron, Estimating long range dependence finite sample properties and confidence intervals, Physica A 312 (2002) 285–299.

J. Bezdek, R. Ehrlich, W. Full, fcm: the fuzzy c-means clustering algorithm, Computers & Geosciences 10 (1984) 191–203.

L. R. Rabiner, A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition. Proceedings of the IEEE, 77 (2), 257-286, 1989.

A. J. Viterbi, Error bounds for convolutional codes and an asymptotically optimum decoding algorithm, IEEE Transactions on Information Theory 13 (2), 260-269. doi:10.1109/TIT.1967.1054010.




DOI (PDF): https://doi.org/10.20508/ijrer.v6i3.4275.g6897

Refbacks

  • There are currently no refbacks.


Online ISSN: 1309-0127

Publisher: Gazi University

IJRER is cited in SCOPUS, EBSCO, WEB of SCIENCE (Clarivate Analytics);

IJRER has been cited in Emerging Sources Citation Index from 2016 in web of science.

WEB of SCIENCE in 2025; 

h=35,

Average citation per item=6.59

Last three Years Impact Factor=(1947+1753+1586)/(146+201+78)=5286/425=12.43

Category Quartile:Q4