Research Of Train Flow Based On Cellular Automata

ABSTRACT:Due to the exponential development of urban rail transport in China, establishing models to simulate and predict the traffic flow is significant in the aspects of railway safe and fastness. This thesis studies the complex traffic flow in the urban rail transporting system with the cellular automata (CA) model. By evaluating the pros and cons for different approaches in the research of urban rail transport, the advantage of CA model in flexibility, extensibility in practice, while retains the complexity and nonlinear characteristics of transporting system are specifically showed in this thesis.Furthermore, the CA model for moving block system running on turning-back operation mode based on the real urban subway system is proposed in this thesis, and the traffic flow model in such operation mode is also researched. Besides, the traffic delay under different operation mode, and its backward propagation phenomenon are researched in this thesis. Additionally, an enhanced model for operation on short route commute is proposed in this thesis, and the characteristics in those models are explored.Main contribution in this thesis:1. Establishes a CA model based on MBS, and plots the space-time diagram for traffic flow. Proposes a more practical model, by considering turning-back operation. Analyzes the influences of train numbers, train interval, dwell time, and turning-back interval on traffic delay employing the enhanced model. The simulation results prove the reliability and accuracy of the model in describing practical MBS.2. Introduce the long/short route operation CA model, based on basic turning-back operation mode CA model. The simulation results prove the model to be accurate and effective by giving the actual behavior of turning-back trains. By simulating the long/short route operation, analyzes the impact on trains’delay under different situations.3. Simulate the traffic flow under mixed route turning-back train operation mode employing the enhanced long/short route operation model. The simulation results show that the short route operation should be arranged first among trains mixed on long/short route. In this way, the whole line would perform more effectively. This result goes well with the fact, and hence could be the theoretical guidance for the real situation.