Power prediction correction method considering photovoltaic power climbing characteristics

The invention belongs to the field of new energy grid-connected consumption, and discloses a photovoltaic power climbing characteristic-considered power prediction correction method, which comprises the following steps of: acquiring historical photovoltaic power data and meteorological factors, performing preprocessing, and extracting photovoltaic power climbing characteristics by utilizing a climbing characteristic calculation formula; establishing an EOF-OPTICS clustering model based on the climbing characteristics to cluster the historical photovoltaic power generation power; performing photovoltaic power prediction on the clustering subset by using an RMSprop-QWGRU photovoltaic prediction model; calculating a photovoltaic power generation power climbing section prediction error; the RMSprop-QWGRU photovoltaic prediction model is used to predict the future photovoltaic power, the ShapeDTW similarity measurement method is used to map the historical photovoltaic prediction error to the future photovoltaic pr