This article presents a comprehensive review of machine learning (ML) techniques for forecasting temperature and solar irradiance in photovoltaic (PV) systems. The study evaluates various ML algorithms, including decision trees, random forests, support vector machines (SVM), and XGBoost, for their effectiveness in predicting weather-related variables crucial for PV performance. The research highlights the importance of accurate forecasting for optimizing grid integration and energy management. Traditional meteorological models face challenges in capturing dynamic weather patterns, making ML a promising alternative due to its ability to learn from historical data and adapt to changing conditions. The study uses a dataset from a solar farm in Hassi R'mel, Algeria, covering seven months of temperature and solar irradiance data. The models are trained on this data to predict future values, with performance evaluated using metrics such as mean absolute error (MAE), mean squared error (MSE), and root mean square error (RMSE). The decision tree model showed the highest accuracy, with minimal error, indicating its effectiveness in capturing temperature patterns. However, the accuracy of ML models depends on the quality of training data and feature selection. The study also discusses the advantages and limitations of each ML method, emphasizing the need for ensemble techniques like XGBoost to improve forecasting accuracy. The results demonstrate that ML models, particularly decision trees, offer a balance between accuracy and computational efficiency, making them suitable for real-time applications in PV systems. The research underscores the importance of integrating advanced ML techniques with climate models to enhance the reliability of solar energy forecasts, contributing to more efficient energy management and sustainable development in renewable energy systems.This article presents a comprehensive review of machine learning (ML) techniques for forecasting temperature and solar irradiance in photovoltaic (PV) systems. The study evaluates various ML algorithms, including decision trees, random forests, support vector machines (SVM), and XGBoost, for their effectiveness in predicting weather-related variables crucial for PV performance. The research highlights the importance of accurate forecasting for optimizing grid integration and energy management. Traditional meteorological models face challenges in capturing dynamic weather patterns, making ML a promising alternative due to its ability to learn from historical data and adapt to changing conditions. The study uses a dataset from a solar farm in Hassi R'mel, Algeria, covering seven months of temperature and solar irradiance data. The models are trained on this data to predict future values, with performance evaluated using metrics such as mean absolute error (MAE), mean squared error (MSE), and root mean square error (RMSE). The decision tree model showed the highest accuracy, with minimal error, indicating its effectiveness in capturing temperature patterns. However, the accuracy of ML models depends on the quality of training data and feature selection. The study also discusses the advantages and limitations of each ML method, emphasizing the need for ensemble techniques like XGBoost to improve forecasting accuracy. The results demonstrate that ML models, particularly decision trees, offer a balance between accuracy and computational efficiency, making them suitable for real-time applications in PV systems. The research underscores the importance of integrating advanced ML techniques with climate models to enhance the reliability of solar energy forecasts, contributing to more efficient energy management and sustainable development in renewable energy systems.