Primjena vještačkih neuronskih mreža za predviđanje snage solarne elektrane
##plugins.themes.bootstrap3.article.sidebar##
##plugins.themes.bootstrap3.article.main##
http://orcid.org/0000-0002-5142-1476
Božidar Popović
http://orcid.org/0009-0009-0397-3776
Aleksandar Anđelković
http://orcid.org/0000-0003-2736-6793
Apstrakt
U radu je predstavljen model za predikciju izlazne snage fotonaponske elektrane korišćenjem vještačkih neuronskih mreža, sa ciljem unapređenja tačnosti prognoze proizvodnje solarne energije. Porast u korišćenju obnovljivih izvora energije, uz nepredvidivost proizvodnje iz solarnih izvora, nameće potrebu za razvojem pouzdanih metoda predviđanja. Analiza je zasnovana na stvarnim podacima prikupljenim u minutnim intervalima, pri čemu su radijacija i temperatura okoline korišćeni kao ulazni parametri, a izmjerena snaga kao izlazni. Model je treniran korišćenjem feedforward vještačke neuronske mreže sa algoritmom Levenberg-Marquardt. Rezultati ukazuju na visok stepen regresije (~0,93), bez prenaučenosti, uz stabilno ponašanje modela i u periodu testiranja. Iako se uočavaju određena odstupanja u ekstremnim uslovima, model se pokazao korisnim za kratkoročno planiranje proizvodnje.
##plugins.themes.bootstrap3.article.details##
Reference
Jobayer, M., Shaikat, M. A. H., Rashid, M. N., Hasan, M. R., A systematic review on predicting PV system parameters using machine learning, Hellyon 9, https://doi.org/10.1016/j.heliyon.2023.e16815, 2023.
Ahmed R., Sreeram V., Mishra Y., Arif, M., A review and evaluation of the state-of-the-art in PV solar power forecasting: Techniques and optimization, Renew. Sustain. Energy Rev., 124, Article 109792, https://doi.org/10.1016/j.rser.2020.109792, 2020.
Marinović B., Kašiković, M., Artificial Neural Networks Models in Solar Photovoltaic and Hydropower Forecasting –a Rreview, Journal of Engineering and Menagement, Vol.3 No. 1, doi: 10.7251/JEM2501035M, 2025.
Rivero-Cacho, A., Sanchez-Barroso, G., Gonzalez-Dominguez, J., Garcia-Sanz-Calcedo, J., Long-term power forecasting of photovoltaic plants using artificial neural networks, Energy Reports, Volume 12, pp 2855-2864, ISSN 2352-4847, https://doi.org/10.1016/j.egyr.2024.08.062, 2024.
Xu, T., Sabzalian, M.H., Hammoud, A. et al., An innovative machine learning based on feed-forward artificial neural network and equilibrium optimization for predicting solar irradiance.Sci Rep14, 2170. https://doi.org/10.1038/s41598-024-52462-0, 2024.
Rahman, S., Rahman, S., Haque, A. K. M. B., Prediction of Solar Radiation Using Artificial Neural Network, Journal of Physics: Conference Series.1767. 012041.https://doi.org/10.1088/1742-6596/1767/1/012041, 2021.
Theocharides, S., Makrides, G., Livera, A., Theristis, M., Kaimakis, P., & Georghiou, G. E., Day-ahead photovoltaic power production forecasting methodology based on machine learning and statistical post-processing. Applied Energy, 268,115023.https://doi:10.1016/j.apenergy.2020.11502, 2020.
Keddouda, A., Ihaddadene, R., Boukhari, A., Atia, A., Arıcı, M., Lebbihiat, N., Ihaddadene, N., Solar photovoltaic power prediction using artificial neural network and multiple regression considering ambient and operating conditions, Energy Conversion and Management, Volume 288, 117-186, https://doi.org/10.1016/j.enconman.2023.117186, 2023.
Fernando, H., Nugroho, S.A., Saryanita, A., Kikumoto, M., Prediction of SPT value based on CPT data and soil physical properties using ANN with and without data normalization, International Journal of Artificial Inteligence Research, Vol. 5, No. 2, pp 123-131, https://doi.org/10.29099/ijair.v5i2.208, 2021.