Optimization Techniques for Solar Photovoltaic Forecasting in IoT‑Driven Smart Energy Management: A Review

Abstract

The presented paper conducts a comprehensive analysis of several Demand Response (DR) programs, advanced devices, and load planning algorithms, offering a detailed assessment based on insights derived from previous studies on Energy Management Systems (EMS). Over the past decade, there has been a noteworthy increase in the generation of electric power from renewable sources, with solar energy experiencing remarkable development in electricity production within the broader landscape of available renewable energy options. Given the pivotal role of solar photovoltaic (PV) energy in ensuring the stability and security of the grid, especially in the context of expansionary monetary policies, the paper underscores the essential nature of reliable PV power generation forecasting. The study references a substantial number of research endeavors that have evaluated PV electricity production from various perspectives. The growing utilization of renewable power is positioned as a strategy to mitigate the adverse environmental impacts of greenhouse gas emissions. Different modeling processes have been explored and adopted in an effort to enhance the predictive accuracy of sustainable energy. The research highlights the increasing demand for renewable energy forecasts in facilitating transmission system supply and demand predictions. Natural events are discussed to elucidate their impact on the performance of PV systems. The paper systematically evaluates photovoltaic power production forecasts, emphasizing the importance of associating incoming data and modeling it before analysis. Various power generation modeling techniques, categorized across multiple classifications, are scrutinized for their effectiveness. The research critically reviews empirical investigations, encompassing quantitative and machine-learning models based on statistical methodologies.