Airspace carbon emission short-term prediction method and system based on graph neural network

The invention discloses an airspace carbon emission short-term prediction method and system based on a graph neural network, and the method comprises the steps: firstly, learning a topological structure of an airspace sector through employing a graph convolutional neural network GCN, and extracting the spatial features of airspace carbon emission data; and then the airspace carbon emission data with the spatial feature information is transmitted to a long short-term memory (LSTM) network, the time change trend of the airspace carbon emission data is learned, the time features of the airspace carbon emission data are extracted, and finally a final prediction result is output through a full connection layer. According to the method, the time dependence and the space dependence of the airspace carbon emission data can be captured at the same time, the capability of describing the spatial-temporal characteristics of the airspace carbon emission data is achieved, and the airspace carbon emission rule of a large-ra