Research On The High-speed Railway Service Plan Rescheduling Based On Network Evolution

China high-speed railway construction is booming in recent years.Xi'an-Chengdu High-speed Railway?Jinan-Qingdao High-speed Railway and other lines were open to traffic and Renaissance EMU was put into use.China high-speed railway has made new achievements in fixed infrastructure and moving equipment.By the end of 2017,the high-speed railway operating mileage has reached 25 thousand kilometers,and the "four vertical and four horizontal" network layout was completed ahead of schedule.Network operating and the increasingly passenger market share make the safety and reliability of high-speed railway acquire considerable attraction.However,high-speed railway system is often hit by unexpected events.Especially under the network operating condition,emergencies not only lead to partial train delay?train cancellation and passengers stranded,but also lead to large-scale delay and a large number of passengers stranded in the whole high-speed railway system because of perturbation propagation through the network.The system presents great instability and vulnerability.The efficient decisions of train operation adjustment in emergency can effectively reduce the interference of emergencies,therefore,service plan rescheduling in emergency becomes a hot issue in the field of railway operation management.Considering the characteristics of high-speed railway network and transportation organization strategy,a novel approach of service plan rescheduling in emergency is proposed from the perspective of complex network evolution in this paper.First of all,based on the complex network theory,this paper builds the high-speed railway geographic network,passenger flow network and train flow network respectively from three aspects:the connection state of high-speed railway facilities,passenger demand and transport service capability.There are close logical interrelations among the three networks.The geographic network is infrastructure constraints for train running,the passenger flow network is demand constraints for service plan formulation,and the train flow network is the topology expression of service plan under the constraints of the geographic network and the passenger flow network.On the basis of the above network model,the topology characteristics of the existing high-speed railway network in China are analyzed in this paper.The geographic network presents the tree network characteristics.The train flow network has the characteristics of large cluster coefficient and small average path length,which presents the small-world network features.Besides,the importance evaluation index of stations is proposed from the view of weighted network topology,which provides a basis for the calculation of station connection degree in the network evolution model.The analysis of the mapping relationship between the train flow network and service plan shows that the two are homogeneous.It provides theoretical support for the reconstructing of train flow network using the network evolution theory and the service plan rescheduling.Secondly,this paper summarizes measures and strategies of railway train operation adjustment under different influence degrees of perturbations.Based on cascading failure theory,the emergency influence propagation model is proposed.According to the degree of influence to trains in the network,trains are divided into 4 categories.Some trains without departing wait at the station before the disrupted section,some departed trains turn back first and then take the detour route,and some departed trains take the detour route directly.There are also some trains without departing can't wait at the station until the capacities are restored.The classification paves the ground to formulate rules for evolution.Thirdly,this paper analyzes the evolution characteristics of the high-speed railway train flow network in emergency.Considering the stop station set and train running time,a K-shortest based path ant colony algorithm is proposed,which is used to search the train detour route in emergency.According to the adjustment principles and strategies of trains in emergency,the service plan rescheduling model based on network evolution is formulated.The model contains three phases:(1)The community set of stations is divided on the basis of the original service plan,and the train routes are determined;(2)The train flow network evolution rules are formulated considering the original service plan,passenger demand,capacity constraints and other factors.The degree of connectivity between stations is determined by factors such as the importance of stations,passenger flow volume,and edge weight and so on,and then the adjusted train flow network is evolved.(3)According to the mapping relationship between train flow network and service plan,we can get the service plan after rescheduling.In addition,the paper sets up the evaluation system of service plan rescheduling from different angles,such as network efficiency,service intensity and train attendance.Finally,this paper takes the Beijing-Shanghai High-speed Railway and the adjacent railway network as an example,builds the emergency scenario and implements the computational test.The result proves the feasibility of the service plan rescheduling method in this paper.The model in this paper is more consistent with the actual situation of high-speed railway operation.It can also make full use of the network capacity.This paper proposes a service plan rescheduling model based on network evolution,it focuses on network operating and increasingly tight relevance among the lines of China high-speed railway.The model is built considering the network factors such as network efficiency,transfer accessibility and traditional factors such as passenger demand,service capability,which is more practical for large-scale network service plan rescheduling than traditional models.