Solar power generation is directly dependent on weather conditions. Temperature, humidity, cloud density, solar radiation and wind speed cause significant changes in PV system performance. But existing studies have typically analyzed one or two climate components separately, leaving an unclear understanding of the combined effects. This results in uncertainty in short-term solar power forecasts, which creates problems in grid planning and energy management. Current research has limitations in analyzing the combined effects of climatic factors and applying machine learning based forecasting models. In particular, the simultaneous analysis of multiple climatic features using real datasets and reliable short-term forecasting with data-driven models has been scarce. This research fills that gap. Analyzing the combined effect of different climatic factors using a real solar dataset and providing reliable PV power forecasting through Random Forest regression model and the model achieved reliable forecasting accuracy (R² ≈ 0.80). This research will help in grid planning, energy determination, and solar plant operations by knowing the short-term solar power forecast.