Under the limitation of the available data of airport surface traffic,it was of great significance for civil aviation transportation system to accurately extract the spatial and temporal characteristics of airport traffic data and predict the surface traffic flow.Firstly,based on the pushback control theory,a numerical simulation model of airport surface operation was established to obtain predictive indicators that couldn't be obtained due to data limitations.Secondly,a combined prediction model of convolutional neural network(CNN)and long short-term memory network(LSTM)was built to extract spatio-temporal features.Finally,taking Henan Zhengzhou Airport as an example,the prediction performance and error index of the model under different training data were compared.The results showed that the prediction model based on simulation index had high accuracy and stable performance.
维普
万方数据
