Forecasting Solar Energy: Leveraging Artificial Intelligence and Machine Learning for Sustainable Energy Solutions
Taraneh Saadati, Burak Barutçu
Abstract
ABSTRACT Integrating solar energy into power grids is essential for advancing a low‐carbon economy, but accurate forecasting remains challenging due to solar output variability. This study comprehensively reviews solar energy forecasting models, focusing on how Artificial Intelligence (AI) and Machine Learning (ML) enhance forecast accuracy. It examines the current landscape of solar forecasting, identifies limitations in existing models, and underscores the need for more adaptable approaches. The primary goals are to analyze the evolution of AI/ML‐based models, assess their strengths and weaknesses, and propose a structured methodology for selecting and implementing AI/ML models tailored to solar power forecasting. Through comparative analysis, the study evaluates individual and hybrid models across different forecasting scenarios, identifying under‐explored research areas. The findings indicate significant improvements in prediction accuracy through AI/ML advancements, aiding grid management and supporting the low‐carbon transition. Ensemble methods, deep learning techniques, and hybrid models show great promise in enhancing reliability. Combining diverse forecasting approaches with advanced AI/ML techniques results in more reliable solar forecasts. The study suggests that improving model accuracy through these integrated methods offers substantial opportunities for further research, contributing to global sustainability efforts, particularly UN SDGs 7 and 13, and promoting economic growth with minimal environmental impact.