Computational Experiments For Emergency Evacuation Of Urban Rail Transit Hub

ABSTRACT:Urban rail transit has been recognized as an important way to trip by virtue of the advantages of safety, high-efficiency, convenience, reliability, energy-saving, low-carbon, etc. The loss and influence will be severe once emergent incident occurs in that the people in urban rail transit hub station is always congested. It’s impossible to carry out frequent drill when the requirements and costing are considered. At the same time, it’s also very hard to analyze or predict the results of emergency evacuation using single mathematical model, because too many factors, including staff, equipments, environment, etc., exist in the emergency evacuation in urban rail transit system. ACP approach, composed of artificial systems, computational experiment, and parallel execution, can provide revisable, repeatable and low-cost computational experiment senses and is a feasible solution to the study of emergency evacuation for urban rail transit. This work establishes emergency evacuation artificial system of typical urban rail transit hub station, related problems in emergency evacuation are studied based on computational experiments, the main contents can be summarized as follows:Firstly, pedestrian social force model for emergency evacuation in urban rail transit hub station is studied, a leader model is proposed and used in the optimization problem in emergency evacuation scheme. Agent-based artificial systems for urban rail transit hub station are developed based on investigate and survey data. Comparing to the traditional social force mode, the leader model considers more factors, including distance between leader and pedestrian, visual angle, excitement level, and so on. Optimal evacuation schemes are obtained based on the computational results of the scenarios of leader-free and leader with different numbers. In view of the fact that leader model performs better in evacuation time than leader-free model, one concludes that the leader model holds the advantages of rapidity, safety and high-efficiency in emergency evacuation.Secondly, by virtue of the adjustability of dynamic emergent identification in urban rail transit hub station, emergency evacuation schemes are optimized based on dynamic identification. Static identification and dynamic identification for urban rail transit hub station are established, which are used to carry out computational experiments at different stages, respectively. Results show that evacuation scheme based on dynamic identification is much more effective to evacuate the pedestrian using information of pedestrian flow and disaster, therefore, mean evacuation time and personal damage are decreased significantly. All in all, dynamic identification can online optimize evacuation scheme for urban rail transit hub station.Finally, passenger flow transmission model for typical station with large passenger flow is proposed, which is adopted to test the station service level and train operation with large passenger flow. By conducting computational experiments with large passenger flow in urban rail transit system, dynamic transmission law of passenger flow is obtained based on the network model of urban rail transit. Further based on the passenger flow transmission model, the effect to station service level and train operation caused by large passenger flow is analyzed, and effective solutions to eliminate the bad effect are proposed, which eventually provide theoretical and practical operation gist for emergency disposition and decision for urban rail transit system.