Weighted least squares method for solar radiation estimation with performance analysis
Journal of Atmospheric and Solar-Terrestrial Physics, cilt.286, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 286
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.jastp.2026.106912
- Dergi Adı: Journal of Atmospheric and Solar-Terrestrial Physics
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Artic & Antarctic Regions, Compendex, INSPEC, Academic Search Ultimate (EBSCO), Engineering Source (EBSCO)
- Anahtar Kelimeler: Model developing, Solar radiation, Statistical analysis, Weighted least squares method
- Çukurova Üniversitesi Adresli: Evet
Özet
Accurate modeling of global solar radiation (GSR) is essential for the efficient design, optimization, and management of renewable energy systems. However, GSR data often exhibit heteroskedastic variance structures that violate the homoskedasticity assumption of Least Squares (LS) regression, leading to inefficient and biased parameter estimations. To overcome this limitation, this study proposes a method based on the Weighted Least Squares (WLS) estimation framework. This method provides efficient parameter estimation for developing a GSR prediction model under unknown heteroskedasticity conditions by applying a two-stage estimation procedure. The performance analysis of the prediction model developed with the WLS method is examined in detail using six different statistical error tests and the results are presented in graphs and table. All calculations in this study were performed using the Matlab program. The results obtained demonstrate that the model developed with the proposed the WLS method offers a statistically rigorous and computationally methodologically robust approach to high-precision GSR estimation, making significant contributions to renewable energy modeling and advanced regression-based prediction systems.