Research And Application Of Spatio-Temporal Path Estimation For Rail Transit Passenger

Under the conditions of urban rail transit network operation,the diversity of passenger travel paths is obvious.Using the actual operation data to accurately restore and estimate the spatio-temporal paths of passengers in the network not only provides data support for the evolution of spatio-temporal distribution of passenger flow and system ticketing,but also provides a basis for the impact assessment of unexpected events,which is of great significance for urban rail operation management.To this end,this paper proposes a spatio-temporal path estimation method for passengers based on spatio-temporal state network,and investigates the impact analysis method of emergency events based on path estimation.The main research contents of this paper include.(1)analyzing the spatio-temporal state characteristics of passenger travel process and passenger travel path of rail transit system,summarizing the characteristics of passenger spatio-temporal path estimation problem under the condition of known local observation data,proposing a research perspective of introducing passenger flow observation data to realize the passenger path estimation problem;using spatio-temporal state network model to realize the dynamic relationship of train flow,passenger flow and key facilities and equipments in rail transit network and other elements The spatiotemporal state network model is used to achieve the dynamic relationship between train flow,passenger flow and key facilities and equipment in the rail transit network.(2)The 0-1 integer planning model for passenger path estimation with the objective of minimizing the passenger path estimation error and generalized cost of passenger travel,and the constraints of network flow balance,facility and equipment capacity and state consistency is established based on the AFC passenger swipe card data and crosssectional flow observation data obtained from train weighing.(3)Based on the generalized least squares approach,the model is decomposed into a passenger path estimation subproblem and a passenger flow state estimation matching subproblem,and the overall model solving algorithm based on rolling update mechanism is proposed based on the coupling relationship between the two decision variables.(4)Based on the qualitative analysis of the impact of the road network under unexpected events,a quantitative assessment method of the impact of unexpected events applying passengers’ individual spatio-temporal paths is proposed,and the specific assessment process and calculation methods of assessment indexes are elaborated;and the passenger spatio-temporal path estimation model is used to realize the retrospective restoration of passengers’ spatio-temporal paths under normal and unexpected event operation scenarios,and then the passenger spatio-temporal path estimation is applied to The results are used to achieve the analysis of the impact of unexpected events.In summary,there are 62 figures,36 tables and 82 references in this paper.