Research On Train Operation Scneduling Model And Algorltnm In Emergency Conditions

ABSTRACT:China has a vast territory, in which the characteristics of railway network is wvide distribution, long lines, mixed passenger and freight trains running, and the complex letwork structure. Importantly, China is one of the frequent natural disasters in the world. In addition, railway transport organizations also face the challenges of public events or other security events, such as sudden large passenger flow. These events can suddenly cause reater negative impact on the order of rail transport organizations in a short time. The abnormal circumstances of network node or network range will pose a threat to the safety d efficiency of rail transport. Train operation dispatching control method under emergency onditions, can largely improve the efficiency of disposal, and reduce the impact of mergency.On the other hand, planning and ordering is the core business of train scheduling command on the condition of complex network. Establishment and adjustment of train operation control program is a combination of large-scale optimization, which involve ransportation, maintenance, electricity, vehicle and many other complex systems business. And emergencies seriously affect the railway capacity and transportation efficiency. Through the theoretical foundation of mathematics, studying the train operation plan adjustment strategy on emergency, has very important academic significance and practical significance. This research can make transport organization achieve the optimum condition quickly and accurately, and reduce the effects of emergencies.Therefore, this article combed the rail system department business relationship and esponsibilities at the disposal of the emergency. And integration of train operation plan and operation scheduling model is established, with these results as boundary constraints. In the rocess of establishing the model, the integrated used algorithms include ant colony algorithm, optimization theory, a multi-objective particle swarm optimization algorithm, genetic algorithm and immune algorithm based on information entropy, etc. This model can rovide the optimal strategy, on the train route allocation and phase plan adjustment.The main research harvests of this dissertation include:(1) This paper introduces Bayesian theory into railway emergency dispatching early warning and control strategy, and set up train operation dispatching early warning and control model, on the basis of the prior probability. Then, this paper calculates the probability set of incident cause, which can provide information support for the train operation dispatching warning on emergency conditions.(2) This paper introduces the concept of road resistance, and improves the ant colony algorithm. On the basis of this, this paper designs the ant colony algorithm introduced into road impedance, to solve the train K-shortest path problem on emergency conditions. In addition, this paper considers K-shortest path algoriththe effected by the train operation stop time under emergency, trying to solve the train scheduling K-shortest path problem on emergency conditions.(3) Based on the analysis of cause of the train route assignment problem, influence factors and constraints, this paper establishes a mathematical model of train route assignment on emergency condition. According to this model, this paper designs the improved differential algorithm for train path assignment problem.(4) Combination of accidents and the train operation control problem in natural disasters, accident and train operation control problem under natural disaster are converted to a multi-objective programming problem, in this paper. In addition, this paper combines multi-objective particle swarm algorithm and genetic algorithm creatively. And on this basis, this paper establish a new hybrid intelligent algorithm based on multi-objective optimization, which can effectively control the operation of the train in accident or under natural disasters.This paper analyzes the train control problem in the situation of major holidays and sudden large passenger flow, and creatively integrates the traditional train operation scheduling and train stops plan of dispatching. This paper establishes a integration model about train operation plan and operation scheduling, and designs an immune algorithm based on information entropy, by adjusting the train stop scheme to deal with sudden large Dassenger flow and major holidays.