Research On Anti-interference Performance And Delay Prediction For Regional Railway Network

The regional railway operation system is an open giant system with complex and changeable operation environment,which is easily affected by emergencies.Due to the difficulty of train delay control,it is urgent to study the anti-interference performance and delay prediction methods of regional railway network under the influence of emergencies.In order to provide decision support for regional railway train operation adjustment,this thesis analyzes the anti-interference performance of regional railway network,clarifies the propagation and dissipation process of regional railway train delays,and predicts the trend of train delays.The main work is as follows:(1)This thesis summarizes the research status at home and abroad,analyzes the connotation and characteristics of regional railway emergencies,designs the measurement index of regional railway network topology structure,gives the evaluation function of regional railway network anti-interference,and qualitatively analyzes the transmission mode and control strategy of regional railway train delay under the influence of emergencies.(2)The topological structure characteristics of regional railway network are explored,and the topological parameters of physical network and order network are considered.In order to clarify the ability of regional railway network to absorb and resist interference under the influence of emergencies,the identification method of key stations of regional railway network based on network fusion is proposed,and the simulation attack algorithm is designed.Then,the anti-interference performance of regional railway network is analyzed through simulation analysis of deliberate and random attack modes.(3)On the basis of clarifying the mechanism of regional railway train delay transmission,combining with the SIRS disease transmission model,delay and self-recovery rate are introduced to construct the regional railway train delay transmission model without delay and considering delay respectively.By making reasonable assumptions about the model parameters,the transmission process of different types of stations as the initial source of infection is simulated to master the delayed transmission process of regional railway network.(4)The distribution characteristics of train delays are analyzed,and the train delay time prediction methods based on RF-LR,RF-MLP and RFXGBoost are designed respectively.Taking the Guiyang-Guangzhou highspeed railway case data as the research object,the prediction effects of the three prediction methods are compared.The results show that the train delay prediction method based on RF-XGBoost has the best effect and can provide decision support for delay control.59 figures,26 tables,97 references are included in this thesis.