Abstract
Photovoltaic (PV) systems generate solar power worldwide. Solar power sources are unpredictable by nature because the output power of PV systems is alternating and heavily dependent on environmental conditions. Among these are wind speed, humidity, PV surface temperature, and irradiance. Planning ahead is essential for solar power generation due to the unpredictable nature of photovoltaic systems, much as forecasting solar electricity is necessary for the electric grid. The irradiance has a significant impact on solar power generation, making weather forecasting challenging and complex. There is discussion of how different environmental factors affect a photovoltaic system's output. In order to overcome the difficulties caused by the variability of solar radiation, this research explores the application of deep learning for photovoltaic (PV) power output prediction. The confusion matrix and ROC AUC results reveal that the proposed deep learning model predicted accurately the power output.