Advertisement

Neural Computing and Applications

, Volume 26, Issue 8, pp 1955–1962 | Cite as

Photovoltaic energy production forecast using support vector regression

  • R. De LeoneEmail author
  • M. Pietrini
  • A. Giovannelli
Original Article

Abstract

Forecasting models for photovoltaic energy production are important tools for managing energy flows. The aim of this study was to accurately predict the energy production of a PV plant in Italy, using a methodology based on support vector machines . The model uses historical data of solar irradiance, environmental temperature and past energy production to predict the PV energy production for the next day with an interval of 15 min. The technique used is based on \(\nu \)-SVR, a support vector regression model where you can choose the number of support vectors. The forecasts of energy production obtained with the proposed methodology are very accurate, with the \(R^{2}\) coefficient exceeding 90 % . The quality of the predicted values strongly depends on the goodness of the weather forecast, and the \(R^{2}\) value decreases if the predictions of irradiance and temperature are not very accurate.

Keywords

Forecasting model Support vector machines PV energy production 

Notes

Acknowledgments

The authors express deep gratitude to the Loccioni Group for providing all the data used in the present research study. In particular to Emanuele Mazzanti for implementing the Photovoltaic Energy Production Model in the web application of the Loccioni Group. A special thanks to the anonymous reviewers for the ideas and suggestions received. The work of the first author was partly supported by the grant INdAM-GNCS Research Project 2014 “Numerical Methods for Nonlinear Optimization”.

References

  1. 1.
    Aznarte J, Girard R, Espinar B, Moussa A, Kariniotakis G (2009) Forecasting functions with focus to PV prediction for microgrids. Tech. rep., project more microgrids. http://www.microgrids.eu/documents/639.pdf
  2. 2.
    Bacher P, Madsen H, Nielsen HA (2009) Online short-term solar power forecasting. Solar Energy 83(10):1772–1783CrossRefGoogle Scholar
  3. 3.
    Bien T, Musikowski HD (2008) Forecasting photovoltaic energy using a fourier series based method. Engineering and system integration, pp 3088–3091Google Scholar
  4. 4.
    Chang CC, Lin CJ (2002) Training v -support vector regression: theory and algorithms. Neural Comput 14(8): 1959–1977 http://dblp.uni-trier.de/db/journals/neco/neco14.html#ChangL02
  5. 5.
    Chang CC, Lin CJ (2011) LIBSVM: A library for support vector machines. ACM Trans Intell Syst Technol 2: 1–27. Software http://www.csie.ntu.edu.tw/~cjlin/libsvm
  6. 6.
    Cococcioni M, D’Andrea E, Lazzerini B (2012) One day-ahead forecasting of energy production in solar photovoltaic installations: an empirical study. Intell Decis Technol 6(3):197–210. doi: 10.3233/IDT-2012-0136 Google Scholar
  7. 7.
    da Silva Fonseca J, Oozeki T, Takashima T, Koshimizu G, Uchida Y, Ogimoto K (2011) Photovoltaic power production forecasts with support vector regression: a study on the forecast horizon. In: Photovoltaic specialists conference (PVSC), 2011 37th IEEE, pp 2579–2583. doi: 10.1109/PVSC.2011.6186475
  8. 8.
    De Cosmis S, De Leone R (2012) Support vector machine for robust regression. In: COMTESSA (ed) Proceedings of the CTW 2012 11th cologne-twente workshop on graph and combinatorial optimization, pp 96–99Google Scholar
  9. 9.
    De Leone R (2011) Support vector regression for time series analysis. In: Operations research proceedings 2010. Springer, Berlin, pp 33–38Google Scholar
  10. 10.
    Drucker H, Burges CJC, Kaufman L, Smola AJ, Vapnik V (1997) Support vector regression machines. In: Mozer MC, Jordan MI, Petsche T (eds) Advances in neural information processing systems 9. MIT Press, Cambridge, pp 155–161Google Scholar
  11. 11.
    Fonseca J Jr, Oozeki T, Takashima T, Koshimizu G, Uchida Y, Ogimoto K (2009) Forecast of power production of a photovoltaic power plant in japan with multilayer perceptron artificial neural networks and support vector machine. Large PV Power Plants and Distributed PV pp 4237–4240Google Scholar
  12. 12.
    Gehler P, Nowozin S (2008) Infinite kernel learning. Technical report 178, Max Planck InstituteGoogle Scholar
  13. 13.
    Gönen M, Alpaydin E (2011) Multiple kernel learning algorithms. J Mach Learn Res 12: 2211–2268. http://dl.acm.org/citation.cfm?id=1953048.2021071
  14. 14.
    Jenkins N, Awad B, Ekanayake JB (2009) Advanced architectures and control concepts for MORE MICROGRIDS. Technical report, more microgrids project consortiumGoogle Scholar
  15. 15.
    Mattera D, Haykin S (1999) Support vector machines for dynamic reconstruction of a chaotic system. In: Schölkopf B, Burges CJC, Smola AJ (eds) Advances in kernel methods. MIT Press, Cambridge, MA, USA, pp 211–241. http://dl.acm.org/citation.cfm?id=299094.299106
  16. 16.
    Müller K-R, Smola AJ, Rätsch G, Schölkopf B, Kohlmorgen J, Vapnik V (1997) Predicting time series with support vector machines. In: Gerstner W et al. (eds) Artificial neural networks—ICANN’97, vol 1327. Springer, Heidelberg, pp 999–1004Google Scholar
  17. 17.
    Ogliari E, Grimaccia F, Leva S, Mussetta M (2013) Hybrid predictive models for accurate forecasting in pv systems. Energies 6(4):1918–1929. doi: 10.3390/en6041918 http://www.mdpi.com/1996-1073/6/4/1918
  18. 18.
    Özögür-Akyüz S, Weber GW (2010) Infinite kernel learning via infinite and semi-infinite programming. Optim Methods Softw 25:937–970. doi: 10.1080/10556780903483349 zbMATHMathSciNetCrossRefGoogle Scholar
  19. 19.
    Özögür-Akyüz S, Weber GW (2010) On numerical optimization theory of infinite kernel learning. J Glob Optim 48(2):215–239. doi: 10.1007/s10898-009-9488-x zbMATHCrossRefGoogle Scholar
  20. 20.
    Özögür-Akyüz S, Ünay D, Smola AJ (2011) Guest editorial: model selection and optimization in machine learning. Mach Learn 85(1–2):1–2CrossRefGoogle Scholar
  21. 21.
    Pellegrini M, De Leone R, Maponi P (2013) Reducing power consumption in hydrometric level sensor network using support vector machines. In: PECCS (ed) Proceedings of the PECCS 2013 international conference on pervasive and embedded computing and communication systems, pp 229–232Google Scholar
  22. 22.
    Schölkopf B, Bartlett P, Smola A, Williamson RC (1999) Shrinking the tube: a new support vector regression algorithm. In: Kearns MS, Solla SA, Cohn DA (eds) Advances in neural information processing systems 11. MIT Press, Cambridge, pp 330–336Google Scholar
  23. 23.
    Schölkopf B, Smola AJ, Williamson RC, Bartlett PL (2000) New support vector algorithms. Neural Comput 12(5):1207–1245. doi: 10.1162/089976600300015565 CrossRefGoogle Scholar
  24. 24.
    Simonov M, Mussetta M, Grimaccia F, Leva S, Zich R (2012) Artificial intelligence forecast of pv plant production for integration in smart energy systems. Int Rev Electr Eng 7:3454–3460Google Scholar
  25. 25.
    Smola AJ, Schölkopf B (2004) A tutorial on support vector regression. Stat Comput 14(3):199–222. doi: 10.1023/B:STCO.0000035301.49549.88 MathSciNetCrossRefGoogle Scholar
  26. 26.
    Stitson M, Gammerman A, Vapnik V, Vovk V, Watkins C, Weston J (1997) Support Vector regression with ANOVA decomposition kernels. http://citeseer.nj.nec.com/stitson97support.html
  27. 27.
    Vapnik V (1982) Estimation of dependences based on empirical data. Springer Series in Statistics, Springer, New YorkzbMATHGoogle Scholar
  28. 28.
    Vapnik VN (1995) The nature of statistical learning theory. Springer, New YorkzbMATHCrossRefGoogle Scholar
  29. 29.
    Vapnik V (1998) Statistical learning theory. Wiley, New YorkzbMATHGoogle Scholar
  30. 30.
    Vapnik V, Chervonenkis A (1964) A note on one class of perceptrons. Automation and remote control 25Google Scholar

Copyright information

© The Natural Computing Applications Forum 2015

Authors and Affiliations

  1. 1.School of Science and TechnologiesUniversity of CamerinoCamerinoItaly
  2. 2.Research@energyLoccioni GroupAngeli di RosoraItaly

Personalised recommendations