Research onAirport Passenger Flow Forecasting Method Based on Combination Model
In order to deal with the security risks caused by the inability to effectively predict the passenger flow of the airport,the decline of service quality and the improper allocation of re-sources and other problems.This paper combines machine learning and deep learning methods to build a combined model of SARIMA-CNN-LSTM to predict airport passenger flow.The model first utilizes SARIMA's ability to process linear and seasonal components of the data to make predictions.However,SARIMA cannot adequately fit nonlinear or complex components,resulting in fitted sequences and residual sequences.However,the effect of using LSTM alone is not good,so the model adopts CNN-LSTM structure to deal with the residual sequence.The structure uses CNN to extract the nonlinear and other complex features of the residual sequence and reduce the dimension to make the sequence adapt to the input requirements of LSTM.The structure uses LSTM model to capture the long-term dependence relationship of the CNN-pro-cessed sequence,and carries out nonlinear modeling and prediction,and obtains the prediction re-sult of CNN-LSTM for residual sequence.Finally,the model combines the predicted results of SARIMA and CNN-LSTM to get the final predicted results.The experimental results show that the combined model constructed in this paper has a good passenger flow prediction effect,which is conducive to the safety and efficiency of airport work.
Airport passenger flow forecastSARIMA modelCNN modelLSTM model
万方数据
维普
