The paper reviews machine learning methods for forecasting solar radiation, emphasizing the importance of accurate predictions for managing renewable energy systems. It highlights the need to predict solar radiation for effective power grid operation and energy management, especially in isolated electrical grids. The review covers various forecasting methodologies, including cloud imagery combined with physical models and machine learning models. Machine learning models, such as neural networks, support vector machines, regression trees, random forests, and gradient boosting, are discussed in detail. The paper also addresses the challenges in evaluating the performance of these models due to the diversity of datasets, time steps, forecasting horizons, and performance indicators. To improve prediction accuracy, hybrid models and ensemble forecasting approaches are proposed. The paper concludes by comparing different machine learning methods and their applications in solar radiation forecasting, noting that neural networks are the most commonly used method.The paper reviews machine learning methods for forecasting solar radiation, emphasizing the importance of accurate predictions for managing renewable energy systems. It highlights the need to predict solar radiation for effective power grid operation and energy management, especially in isolated electrical grids. The review covers various forecasting methodologies, including cloud imagery combined with physical models and machine learning models. Machine learning models, such as neural networks, support vector machines, regression trees, random forests, and gradient boosting, are discussed in detail. The paper also addresses the challenges in evaluating the performance of these models due to the diversity of datasets, time steps, forecasting horizons, and performance indicators. To improve prediction accuracy, hybrid models and ensemble forecasting approaches are proposed. The paper concludes by comparing different machine learning methods and their applications in solar radiation forecasting, noting that neural networks are the most commonly used method.