Forecasting of Solar Radiation from Meteorological Data by Multiple Linear Regression Analysis in Mueang Phitsanulok District (Thailand)

Abstract

This research article aims to study the prediction of the bi-weekly solar radiation in Muang Phitsanulok district by constructing predictive equations by multiple linear regression analysis using independent variables consisting of maximum air temperature and average daytime air temperature. The information can be obtained from general meteorological websites. The information is divided into 2 parts. The first part, 72 pieces of information, was collected from January 2012 to December 2014. It was used to construct predictive equations by multiple linear regression analysis. The second part, 63 pieces of information, was collected from January 2015 to September 2017.  It was used to test the prediction results obtained from the predictive equation. The statistics used in this study were the mean absolute percentage error (MAPE). The results of the study showed that MAPE is 6.134%.