AI-Based Solar Power Forecasting for Utility-Scale Photovoltaic Plants: A Review

Abstract

Solar photovoltaic (PV) energy has experienced substantial expansion in electricity generation, with a considerable number of PV systems integrated into grid-connected networks in recent years. However, solar PV power generation is highly intermittent and volatile due to its reliance on solar irradiance and other weather conditions. This variability poses challenges to power systems and become a major constraint on the operation of utility-scale PV plants. As a result, accurate forecasting and effective operational strategies are essential for ensuring the safe and stable integration of utility-scale PV system into the grid. Research on Artificial Intelligence (AI) algorithms, such as machine learning, artificial neural networks and deep learning has been widely reported in different perspectives.  Nevertheless, forecasting research on utility-scale PV system applications remains limited compared with research on small-scale PV systems. This paper reviews AI-based forecasting methods for utility-scale PV plants, highlights key forecasting challenges and national grid code requirements, aiming to support future development of reliable solar forecasting solutions.