APPLICATION OF NEURAL NETWORK TECHNOLOGIES 
TO CALCULATE TOTAL SOLAR RADIATION FLUXES

Nikolaev Aleksandr Anatolevich, Candidate of geographical sciences, associate professor, sub-department of meteorology, climatology and atmosphere ecology, Kazan (Volga region)  Federal University (18 Kremlyovskaya street, Kazan, Russia), Aleksandr.Nikolaev@kpfu.ru
Ismagilov Nail Vagizovich, Candidate of geographical sciences, associate professor, sub-department of meteorology, climatology and atmosphere ecology, Kazan (Volga region) Federal University (18 Kremlyovskaya street, Kazan, Russia), Nail.Ismagilov@kpfu.ru

551.5

10.21685/2307-9150-2017-3-7

Background. Solar radiation is the main factor in many physical, chemical and biological processes on the earth’s surface. However, solar flux data are not always available for a variety of reasons, e.g. a lack of meteorological stations or solar radiation observations on them. Thus, calculation of solar radiation fluxes appears to be a topical question. The main purpose of this study is to review artificial neural networks (ANN), in order to identify suitable models for calculation of solar radiation and to reveal research gaps.
Materials and methods. The data from the Meteorological Observatory of Kazan Federal University were taken as the initial research material.
Results. Studying of literary resources indicates that artificial neural network methods of solar radiation prediction are more accurate compared to conventional methods. Ten models with different input parameters have been developed and the best of them have been revealed.
Conclusions. The results indicate that the use of neural network models for calculation of solar radiation fluxes is a promising area of research. These models allow us to calculate solar radiation characteristics with the use of meteorological parameters. 

neural networks, total radiation, prediction

1. Khabutdinov Yu. G., Shantalinskiy K. M., Nikolaev A. A. Uchenie ob atmosfere: ucheb. posobie [The teaching about the atmosphere: teaching aid]. Kazan: Kazanskiy gos. un-t, 2010, 245 p.
2. Perevedentsev Yu. P., Nikolaev A. A. Klimaticheskie resursy solnechnoy radiatsii i vetra na territorii Srednego Povolzh'ya i vozmozhnosti ikh ispol'zovaniya v energetike [Climatic resources of solar radiation and wind in the territory of Middle Volga region and their application opportunities in power engineering]. Kazan: Otechestvo, 2002, 122 p.
3. Nikolaev A. A. Vestnik Udmurtskogo universiteta. Ser. 6, Biologiya. Nauki o Zemle [Bulletin of Udmurtia University. Series 6. Biology. Geosciences]. 2013, iss. 1, pp. 130–135.
4. Neyronnye seti. STATISTICA Neural Networks: Metodologiya i tekhnologii sovremennogo analiza dannykh [Neural networks. STATISTICA Neural Networks. Methodology and technologies of contemporary data analysis]. Ed. by V. P. Borovikov. 2nd ed., revised and supplemented. Moscow: Goryachaya liniya – Telekom, 2008, 392 p.
5. Sharifi S. S., Rezaverdinejad V., Nourani V. Journal of Atmospheric and Solar-Terrestrial Physics. 2016, vol. 149, pp. 131–145.
6. Besharat Fariba, Dehghan Ali A., Faghih Ahmad R. Renewable and Sustainable Energy Reviews. 2013, vol. 21, pp. 798–821.
7. Rehman S., Mohandes M. Energy Policy. 2008, vol. 63, pp. 571–576.
8. Linares-Rodríguez A., Ruiz-Arias J., Pozo-Vázquez D., Tovar-Pescador J. Energy. 2011, vol. 36, pp. 5356–5365.
9. Koca A., Oztop H. F., Varol Y., Koca G. O. Expert Syst Appl. 2011, vol. 38, pp. 8756–8762.
10. Khatib T., Mohamed A., Sopian K., Mahmoud M. Int J Photoenergy. 2012, pp. 1–16. Available at: www.hindawi.com
11. Elminir H. K., Areed F. F., Elsayed T. S. Sol Energy. 2005, vol. 79, pp. 270–279.
12. Tymvios F. S., Jacovides C. P., Michaelides S. C., Scouteli C. Sol Energy. 2005, vol. 78, pp. 752–762.