Photovoltaic power generation power prediction method and device based on multi-input model

The invention relates to the technical field of new energy power generation prediction, and discloses a photovoltaic power generation power prediction method and device based on a multi-input model, and the method comprises the steps: carrying out the preprocessing of cloud picture data and meteorological data corresponding to photovoltaic; a constructed multi-input model inputs the preprocessed cloud picture data and meteorological data into two different feature processing branch networks to be processed to obtain corresponding features, the features are fused through a feature fusion layer, the fused features are processed into new features through a sharing hidden layer, and the new features are processed into the cloud picture data and the meteorological data. Photovoltaic power generation power prediction is carried out on the input and output layers; training the multi-input model to obtain a trained model as a prediction model; and inputting cloud picture data and meteorological data corresponding to