Knowledge and data fusion driving-based photovoltaic output prediction method

The invention discloses a photovoltaic output prediction method based on knowledge and data fusion driving, and relates to the technical field of power systems, and the method comprises the steps: collecting photovoltaic historical data, geographic position data and NWP weather data, carrying out the data preprocessing, analyzing the mechanism of photovoltaic panel absorption irradiance, building an effective irradiance mechanism model, and carrying out the prediction of the photovoltaic output. According to power generation data and weather data, the data are clustered into three scenes of sunny days, cloudy days and rainy days, a feature sequence is obtained by combining the weather data and a mechanism model, a CNN-LSTM neural network is used for modeling in different weather scenes to obtain photovoltaic output prediction models in different scenes, and weather categories in a period of time in the future are judged. A corresponding algorithm model is input to obtain photovoltaic output prediction data, i