Airspace traffic flow prediction method based on attention space-time diagram convolutional network

The invention discloses an attention space-time diagram convolutional network-based airspace traffic flow prediction method. The method comprises the following steps of: (1) acquiring air traffic flow data and preprocessing the air traffic flow data; step (2), carrying out airspace sector network modeling; (3) constructing an air traffic flow time sequence; step (4), constructing an attention space-time diagram convolutional network; step (5), training the attention space-time diagram convolutional network; step (6), testing the attention space-time diagram convolutional network; aiming at a complex airspace traffic flow prediction problem, the method can simultaneously capture the spatial dependence and the time dependence of the air traffic flow, and has the capability of describing the spatial-temporal characteristics of the air traffic flow, and the attention mechanism can capture the global time dynamic trend of the air traffic flow. The air traffic flow operation rule of a large-range airspace scale can