Abstract: Photovoltaic power generation has intermittent nature and volatility. The accurate prediction of photovoltaic power generation is an important measure to reasonably arrange the operation mode and response measures, and improve the safety and reliability of the power grid. The data-driven method is the main technical means in the research of photovoltaic power generation prediction. However, researchers are faced with problems such as difficulty of obtaining actual data, poor data quality and insufficient data volume. The public datasets of photovoltaic power generation provide researchers with data resources and the basis for comparative analysis of various prediction methods. In this paper, the current situation of the existing public datasets for solar photovoltaic power generation forecasting at domestic and abroad is summarized. The features and advantages and disadvantages of each datasets are sorted out. Finally, suggestions are presented to promote the development of the application of open datasets.