Comparison of ANN Global Horizontal Irradiation predictions with Satellite Global Horizontal Irradiation using Statistical evaluation

Abstract

The most important factor to take into account when building solar energy systems is solar irradiation. It is impossible to measure sun irradiation everywhere due to its high cost and difficulties. Additionally, in some places, the GHI was overpredicted by 25% by NASA satellite data. The main goal of this study was to develop an artificial neural network (ANN) model that can reduce the error in satellite data by predicting global horizontal irradiation (GHI) using inputs from satellite data obtained from the NASA Power Data viewer. The MAPE in the satellite was decreased by 35.8% in Peshawar, 10.2% in Islamabad, and 8.9% in Multan using the ANN models. Additionally, the results showed that all ANN models' predictions were more precise than satellite data.