Research On Train Operation And Dispatching Analysis System Based On Record Data

With the continuous expansion of railway network and the gradual improvement of infrastructure construction in China,travelling by train has become the preferred choice for more and more passengers.As a result,it leads to the distribution mismatch between passenger flow and transport capacity,as well as dispatching problems in passenger transportation,which further contributes to the more frequent adjustments of the train operation diagram.Therefore,it is necessary to analyze the implementation of the train operation diagram based on the record data,so as to grasp the occurrence rule of the deviation of the actual operation diagram from the planned operation diagram,which can not only provide a reference for the dispatchers,but also provide a reliable basis for the adjustment of the next operation diagram.As the related technologies of data mining has matured in recent years,and has achieved certain application results in different disciplines,it is of great value and significance to explore its application in railway transportation organization.This thesis mainly studies the application of data mining technology in improving the redemption rate of train operation diagram.Through the data visualization technology,the existing performance analysis of the train operation diagram is improved,then the delay regularity is mined through association rules,which provides certain decision support and technical ideas for the dispatching work and the operation diagram adjustment.Firstly,through the analysis and summary of the research and development of the methods of record data analysis,data visualization technology and association rule technology at home and abroad,the experience and ideas of previous studies are concluded,and the significance and possibility of the application of visualization technology and association rules in record data are proposed.Secondly,this thesis briefly introduces the visualization technology and its application direction,as well as the implementation process of association rules,its related algorithms and improvement ideas,and summarizes the train operation diagram and the reasons for the actual train operation deviating from the planned train operation diagram,which provides a theoretical basis for the application of data visualization technology and association rule technology in record data analysis.Thirdly,the multi-angle statistical analysis of the execution of the train operation diagram is carried out.And by choosing appropriate data visualization tools and combining interactive visualization with static visualization,the spatial-temporal distribution characteristics of train delays are depicted,so as to enhance the expressiveness of the implementation of the operation diagram.Next,taking the relevant facts before the train arrives at a station as a transaction,10 attributes related to delays are selected to preprocess the performance data.And aiming at the unsatisfactory execution efficiency of Apriori algorithm and combining with practical application requirements,the improved idea of the algorithm is put forward.Then,according to the association analysis between different attributes and attribute combinations and delays,multiple sets of strong association rules with guiding significance for practice are obtained.Finally,combined with the application and analysis of visualization technology and association rule technology,the train operation dispatching analysis system is proposed,and its structure and function are designed and explained according to the target requirements.