Determining the Optimal Empirical Model for Estimating Global Solar Radiation in the Eastern Mid-Hills of Dhankuta, Nepal

Abstract

Accurate estimation of global solar radiation (GSR) is essential for designing, sizing, and evaluating the performance of solar energy systems. However, in Nepal, direct solar radiation data are limited due to the high cost of advanced measuring instruments. To address this gap, the present study develops and validates empirical models to estimate daily average GSR using easily measurable parameters such as sunshine duration, temperature, and relative humidity. Meteorological data for Dhankuta (26.983°N, 87.346°E, 1192 m altitude) were collected for 2021–2022, including sunshine hours, air temperature, atmospheric pressure, wind speed, and relative humidity. Fifteen mathematical models were tested, and their performance was evaluated using four widely accepted statistical indicators: Mean Bias Error (MBE), Mean Percentage Error (MPE), Root Mean Square Error (RMSE), and Coefficient of Determination (R2 ). Among the evaluated models, Model 10 (a modified form of the Abdalla model) achieved the best statistical performance, with the lowest MBE, RMSE, and MPE values and the highest R2 (0.637). The derived empirical constants for Model 10 are: a = –0.1976, b = 0.37824, c = 0.0209, d = –0.00057. The goal of this study is twofold: first, to identify the statistically optimal model, and second, to assess its broader applicability in Dhankuta and similar geographic regions of Nepal. This dual perspective ensures both methodological rigor and practical value for engineers, architects, agriculturists, hydrologists, and policymakers engaged in solar energy planning and applications.