Meteorological factors and their effects on the prediction of the Solar Global Radiation amount in Oman is the focus of this research study. The essence of the investigation was to provide accurate predictions to improve solar energy application effectiveness. The research utilized data from January 2013 through December 2024, and advanced analytical tools, in-cluding heatmaps, box plots, and time-series analysis were utilized to explore the relationship of temperature, rainfall, and solar global radiation. The conclusions of this research featured a moderate positive correlation between the temperature and solar global radiation, while rain-fall was found to have an inverse relationship with temperature. The study also noted seasonal correlations of solar global radiation; the highest solar global radiation readings were about 800 watts per square meter (W/m²) during the summer months, and the lowest were record-ed during the winter months. The study also found that extreme weather events contribute to fluctuations in solar radiation, highlighting the need for reliable and accurate forecasting models in this field. In addition to providing insight into meteorological factors, the study also conducted an evaluation of machine learning models, including Long Short-Term Memory (LSTM) networks, Decision Trees, and Ensemble Extra Trees, to develop improved forecasting accuracy for these variables. The study demonstrated that the LSTM model configured with Particle Swarm Optimization achieved the best forecasting accuracy metrics; Mean Absolute Error (MAE) of 19.978, Root Mean Square Error (RMSE) of 20.930, and R² value of 0.974, indicate an accurate level of prediction reliability. This research presents a novel approach for optimizing Long Short-Term Memory (LSTM) networks. The main contribution of this research, therefore, is to use the findings from this study to create better forecasting models for solar radiation and ultimately provide tools to improve climate modeling, renewable en-ergy planning, and sustainability. Improved forecasting improves the ability to integrate more solar energy into existing grids (i.e., grid stability) and creates opportunities for developing climate-resilient energy solutions.