Solar radiation forecast with multivariate Ridge regression and multivariate Lasso regression methods supported by extraterrestrial radiation

Öz

In order to provide an accurate and reliable energy projection, it is of the utmost importance to ascertain the success of the prediction models to be created using historical data. In this context, in order to determine the electrical energy to be obtained from solar energy, it is essential to estimate the amount of solar radiation on the solar panel with high accuracy. In this study, hourly solar irradiance estimation was performed using MRR and MLR models with input variables generated in three different formats. Among the three formats, designated as V1, V2, and V3, which employ binary and ternary combinations of irradiance, temperature, and relative humidity datasets as input data, the MRR-V1 model was found to yield the most successful prediction result (RMSE=39.76 W/m2). Furthermore, it was observed that the results filtered with ERF exhibited a slight improvement (RMSE=38.42 W/m2).

Anahtar Kelimeler

• Solar Radiation Forecast
• Deep Learning
• Multivariate Ridge Regression
• Multivariate Lasso Regression

Dünya dışı ışınım destekli çok değişkenli Ridge ve Lasso regresyon yöntemleri ile güneş ışınımı tahmini

Öz

Doğru ve güvenilir bir enerji projeksiyonu sunabilmek adına geçmiş veriler kullanılarak oluşturulacak tahmin modellerinin başarısı oldukça önemlidir. Bu bağlamda, güneş enerjisinden elde edilecek elektrik enerjisini belirleyebilmek adına en temelde güneş paneli üzerine gelen güneş ışınımı miktarının yüksek doğrulukta tahmin edilmesi gerekmektedir. Bu çalışmada, üç farklı formatta oluşturulan giriş değişkenleri MRR ve MLR modelleri kullanılarak saatlik güneş ışınımı tahmini gerçekleştirilmiştir. V1, V2 ve V3 ismi verilen ve giriş verisi olarak ışınım, sıcaklık ve bağıl nem veri setlerinin ikili ve üçlü kombinasyonları kullanılan bu formatlar arasında en başarılı tahmin sonucunu MRR-V1 modelinin verdiği tespit edilmiştir (RMSE=39.76 W/m2). Son olarak, ERF ile filtrelenen sonuçların düşük bir miktar daha iyileştiği görülmüştür (RMSE=38.42 W/m2).

Anahtar Kelimeler

• Güneş Işınım Tahmini
• Derin Öğrenme
• Çok Değişkenli Ridge Regresyon
• Çok Değişkenli Lasso Regresyon

Kaynakça

1. 1. Wu J., Chan C.K., Prediction of hourly solar radiation using a novel hybrid model of ARMA and TDNN, Solar Energy, 85 (5), 808-817, 2011.
2. 2. Voyant C., Randimbivololona P., Nivet M.L., Paoli C., Muselli M., Twenty four hours ahead global irradiation forecasting using multi-layer perceptron, Meteorological Applications, 21, 644-655, 2014.
3. 3. Ismail M.T., Abu Shah N.Z., Abdul Karim S.A., Modeling solar radiaiton in Peninsular Malaysia using ARIMA model, Clean Energy Opportunities in Tropical Countries, 53-71, 2021.
4. 4. Shadab A., Ahmad S., Said S., Spatial forecasting of solar radiation using ARIMA model, Remote Sensing Applications: Society and Environment, 20, 2020.
5. 5. Mukhtar M., Oluwasanmi A., Yimen N., Qinxiu Z., Ukwuoma C.C., Ezurike B., Bamisile O., Development and comparison of two novel hybrid neural network models for hourly solar radiation prediction, Applied Sciences, 12 (3), 2022.
6. 6. Eşlik A.H., Şen O., Serttaş F., CNN-LSTM model for solar prediction: performance analysis, Journal of the Faculty of Engineering and Architecture of Gazi University, 39 (4), 2155-2162, 2024.
7. 7. Sansa I., Boussaada Z., Bellaaj N.M., Solar radiation prediction using a novel hybrid model of ARMA and NARX, Energies, 14 (21), 2021.
8. 8. Akarslan E., Hocaoğlu F.O., Edizkan R., Novel short term solar irradiance forecasting models, Renewable Energy, 123, 58-66, 2018.

9. 9. Gürlek C., Şahin M., Estimation of the global solar radiaiton with the artificial neural networks for the city of Sivas, European Mechanical Science, 2 (2), 46-51, 2018.
10. 10. Arseven B., Çınar S.M., Dünya dışı ışınımlarla iyileştirilmiş ARIMA, ridge regresyon ve lasso regresyon yöntemlerinin saatlik güneş ışınım tahmininde kullanılması, Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi, 12 (3), 733-742, 2023.
11. 11. Kızıltan Ö., Şahin M., Calculation of Solar Radiation by Using Regression Methods, International Physics Conference at the Anatolian Peak (IPCAP2016), Erzurum-Türkiye, 25-27 Şubat, 2016.
12. 12. Mirzabe A.H., Hajiahmad A., Keyhani A., Mirzaei N., Approximation of daily solar radiation: A comprehensive review on employing of regression models, Renewable Energy Focus, 41, 143-159, 2022.
13. 13. Bouchouicha K., Hassan M.A., Baailek N., Aoun N., Estimating the global solar irradiation and optimizing the error estimates under Algerian desert climate, Renewable Energy, 139, 844-858, 2019.
14. 14. Yao W., Li Z., Wang Y., Jiang F., Hu L., Evaluation of global solar radiation models for Shanghai, China, Energy Conversion and Management, 84, 597-612, 2014.
15. 15. Yıldırım H.B., Çelik Ö., Teke A., Barutçu B., Estimating daily global solar radiation with graphical user interface in Eastern Mediterranean region of Turkey, Renewable and Sustainable Energy Reviews, 82, 1528-1537, 2018.
16. 16. Rensheng C., Shihua L., Ersi K., Jianping Y., Xibin J., Estimating daily global radiation using two types of revised models in China, Energy Conversion and Management, 47 (7-8), 865-878, 2006.
17. 17. Rao K.D.V.S.K., Premalatha M., Naveen C., Analysis of different combinations of meteorological parameters in predicting the horizontal global solar radiation with ANN approach: A case stuıdy, Renewable and Sustainable Energy Reviews, 91, 248-258, 2018.
18. 18. Abdalla Y.A.G., New correlations of global solar radiation with meteorological parameters for Bahrain, International Journal of Solar Energy, 16 (2), 111-120, 1994.
19. 19. Chen R., Ersi K., Yang J., Lu S., Zhao W., Validation of five global radiation models with measured daily data in China, Energy Conversion and Management, 45 (11-12), 1759-1769, 2004.
20. 20. Yıldırım H.B., Teke A., Torres F.A., Evaluation of classical parametric models for estimating solar radiation in the Eastern Mediterranean region of Turkey, Renewable and Sustainable Energy Reviews, 82, 2053-2065, 2018.
21. 21. Trabea A.A., Shaltout M.A.M., Correlation of global solar radiation with meteorological parameters over Egypt, Renewable Energy, 21 (2), 297-308, 2000.
22. 22. Meteoroloji Genel Müdürlüğü. Resmi istatistikler. https://www.mgm.gov.tr/veridegerlendirme/il-ve-ilceler-istatistik.aspx. Erişim tarihi Haziran 6, 2024. 23. Hoerl A.E., Kennard R.W., Ridge regression: Biased estimation for nonorthogonal problems, Technometrics, 42 (1), 80-86, 2000.
23. 24. Marquardt D.W., Snee R.D., Ridge regression in practice, The American Statistician, 29 (1), 3-20, 1975.
24. 25. Topal M., Eyduran E., Yağanoğlu A.M., Sönmez A.Y., Keskin S., Çoklu doğrusal bağlantı durumunda ridge ve temel bileşenler regresyon analiz yöntemlerinin kullanımı, Atatürk Üniversitesi Ziraat Fakültesi Dergisi, 41 (1), 53-57, 2010.
25. 26. Sandia National Laboratories. PV performance modeling collaborative (PVPMC). https://pvpmc.sandia.gov/modeling-guide/1-weather-design-inputs/irradiance insolation/extraterrestrial-radiation/. Erişim tarihi Haziran 6, 2024.
26. 27. Fixsen D., Blase K., Metz A., Dyke M.V., Implementation Science, International Encyclopedia of the Social & Behavioral Sciences, FPG Child Development Institute, University of North Carolina at Chapel Hill, Chapel Hill, NC, A.B.D., 695-702, 2015.
27. 28. James G., Witten D., Hastie D., Tibshirani R., An Introduction to Statistical Learning, Springer, Heidelberg, Almanya, 2013.
28. 29. Taylor K.E., Summarizing multiple aspects of model performance in a single diagram, Journal of Geophysical Research Atmospheres, 106, 7183-7192, 2001.
29. 30. Saleh A.K.M.E., Arashi M., Kibria B.M.G., Theory of Ridge Regression Estimators with Applications, Wiley, New Jersey, A.B.D., 2019.