Research On Civil Aviation Flight Path Clustering Method Based On Trajectory Data Mining And Deep Learning

Aircraft trajectory data analysis technology can obtain a lot of useful information from the historical data of aircraft trajectory.It plays an important role in airspace management,civil aviation flight scheduling,Airport management,aviation enterprise fuel cost management and so on.At present,data mining technology is used to cluster the analysis of aviation trajectory data.The selection of clustering method and the extraction of sample data are the main factors that affect the accuracy of clustering analysis.Firstly,this paper analyses the clustering analysis methods commonly used in trajectory analysis.Secondly,in order to establish a hybrid framework for the analysis of Aeronautical trajectory data,several typical deep learning algorithms,such as LSTM and GRU,are deeply studied,and applied to feature extraction of Aeronautical trajectory data.A specific experimental scheme is designed to realize the learning algorithms of single-layer and multi-layer networks.Experimental comparison method is used to discuss the learning effect of the algorithm.Finally,referring to data processing of ship navigation trajectory? [1] The analysis framework,which combines the deep learning method with the clustering analysis method,proposes a framework suitable for the analysis of air trajectory data,and designs an experimental scheme to verify the effectiveness of this analysis framework in improving the analysis of air trajectory data.The experimental results based on real flight trajectory data show that the accuracy of clustering analysis can be significantly improved by using sliding window algorithm to annotate the characteristics of flight trajectory data.In clustering analysis,if the deep learning method is used to extract the features of the trajectory data first,the accuracy of the subsequent clustering analysis will be significantly affected.The combination of the deep learning method and the clustering method can effectively improve the accuracy of the airline trajectory data analysis.