STUDY OF VARIABLES AFFECTING THE PREDICTED SOLAR RADIATION INTENSITY BY ANN METHOD: CASE STUDY IN MUEANG PHITSANULOK DISTRICT

Abstract

This research article is a study of the variables affecting the predicted solar radiation intensity using artificial neural networks (ANN), case study in Mueang Phitsanulok district. The objective of this study is to be used as a guideline for creating basic information that leads to building confidence for entrepreneurs in the establishment of solar power plants in the area of Mueang Phitsanulok district. The variables used in the study consisted of Altitude, Azimuth, Declination, Eccentricity Correction Factor, Hour Angle, Mass Air, the solar radiation transmission coefficient caused by solar radiation scattering by air molecules, Angstrom's turbidity coefficient, Cloud cover and Maximum Ambient Temperature. The study was divided into two parts: predicting the solar radiation intensity in each variable and predicting the solar radiation intensity from the variable group. The results of the study showed that the single variable that has the most effect on predicting maximum solar radiation intensity is the maximum ambient temperature with the correlation value 0.949. The variable group that has the most effect on predicting the solar radiation intensity is the variable groups consisting of Altitude, Azimuth, Declination, Eccentricity Correction Factor and Hour Angle with an average percentage error of 4.383 percent.