The invention relates to a single-step progressive type photovoltaic power station ultra-short-term power prediction method and system. The method includes the following steps that: the meteorologicaldata of a prediction date are obtained through using a pre-established ARIMA time series model, wherein the ARIMA time series model is acquired based on historical meteorological data; the output power of a photovoltaic power station on the prediction date is predicted by using a pre-established SVR regression model according to the meteorological data of the prediction date, wherein the SVR regression model is acquired based on the historical meteorological data and the output power of the photovoltaic power station corresponding to the historical meteorological data; and with the initialized historical meteorological data adopted as an independent variable, and the output power of the photovoltaic power station corresponding to the initialized historical meteorological data as a dependent variable, an SVR regression model for predicting the output power of the photovoltaic power station is trained. According to the single-step progressive type photovoltaic power station ultra-short-term power prediction method and system of the invention provided by the technical schemes of the present invention adopted, the output power is predicted through the SVR regression model and by meansof the meteorological data with large correlation with the output power of the photovoltaic power station.