Modelling of sunshine estimation using regression analysis based on data mining: A case study in Sakarya Basin


Yuksel İ., Sandalci M.

Physics and Chemistry of the Earth, cilt.136, 2024 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 136
  • Basım Tarihi: 2024
  • Doi Numarası: 10.1016/j.pce.2024.103771
  • Dergi Adı: Physics and Chemistry of the Earth
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, Aquatic Science & Fisheries Abstracts (ASFA), Artic & Antarctic Regions, Chimica, Communication Abstracts, Compendex, Geobase, INSPEC, Metadex, Pollution Abstracts, Civil Engineering Abstracts
  • Anahtar Kelimeler: Data mining, Energy consumption, Energy generation, Hydroelectric, Time series, Turkey
  • Yıldız Teknik Üniversitesi Adresli: Hayır

Özet

Knowing the Duration of Sunshine (Shs) is very vital in agricultural production, utilization of solar energy and similar issues. However, determination of sunshine by using direct measurement is laborious, time consuming, and costly. Therefore, it can be estimated by using other approaches instead of direct measurement. Accordingly, in order to estimate sunshine hours, some climate data which have been obtained from Sakarya meteorological station such as observed Duration of Sunshine (Shs), Maximum Temperature (Tmax), Minimum Temperature (Tmin), Average Temperature (Tave), Monthly Evaporation (Em), Relative Humidity (Rh) and precipitation height (P) are needed. In this study, climate data were analyzed by using simple and multiple linear regression analysis and the Shs were estimated. Firstly, normal distributions of climate data sets were examined by using Kolmogorov-Simirnov test and descriptive statistical values. Secondly, Pearson coefficient, which is a parametric method, was determined for the correlation of the climate data. Thirdly, models were formed by simple and multiple regression and sunshine was estimated. Later, the predictions which have been obtained from the models are compared with the observation values and determination Coefficients of Determination (R2) are found. At the end, just to verify the predicted and observed sunshine data sets, the performance of their criteria such as Mean Absolute Error (MAE), Mean Square Error (MSE) and R2 were compared with each other. Because of the study, it was found that in the estimation of sunshine, Model 6 determined the best results.