CONCERNING THE MATTER OF THE (IM)PRACTICALITY OF SOLAR FORECASTING MODELS

Abstract

The availability of solar radiation data is crucial for determining the appropriate sizing of solar energy systems. As solar energy is widely used in electricity generation, numerous research efforts have been dedicated to developing models capable of estimating solar irradiance from various perspectives. These prediction models can be categorized as satellite-based, regression-based, statistical, artificial intelligence-driven, or hybrid in nature. While significant progress has been made during the development of these models, there are concerns among some researchers regarding their practical applicability. In this study, we aim to provide a comprehensive overview of existing solar irradiance prediction models and conduct a prospective and critical analysis of their practicality and accessibility based on the existing literature. Special emphasis is placed on the importance of researchers meticulously studying the current gaps in research and actively working to enhance and implement promising studies to overcome any shortcomings in the prediction models. It is worth noting that generalizing solar irradiance prediction models for locations without direct measuring instruments poses a challenging task. Thus, this article contributes valuable insights to researchers, practitioners, investors, and all stakeholders interested in advancing and utilizing solar irradiance prediction models to support the development of efficient solar energy systems. By shedding light on the strengths and weaknesses of existing models, we aim to facilitate more accurate and reliable solar energy estimations, thereby encouraging the broader adoption of sustainable and renewable energy sources. Ultimately, this research seeks to foster the growth and successful implementation of solar energy systems on a global scale.