In the literature, there are many different photovoltaic (PV) component sizing methodologies, including the PV/inverter power sizing ratio, recommendations, and third-party field tests. This study presents the state-of-the-art for gathering pertinent global data on the size ratio and provides a novel inverter sizing method. The size ratio has been noted in the literature as playing a significant role in both reducing power clipping and achieving system optimization. The majority of researchers observed that due to varying irradiance distributions and operating temperatures at particular sites, the sizing ratios were dependent on geographic latitude. This study will identify the issue that makes it challenging to acquire dependable and optimum performance for the use of grid-connected PV systems by summarizing the power sizing ratio, related derating factor, and sizing formulae approach. The present study recommends a Deep Learning technique that might, due to the dynamic behavior of the PV technologies, provide fully automatic computation for the DC/AC sizing ratio, and effectively lower the whole return on investment (ROI) over a variety of circumstances and climatic changes.