基于元胞自动机的单车道交通流模型研究

发布时间：2018-05-29 17:59

本文选题：交通流模型 + 元胞自动机　；参考：《长安大学》2014年硕士论文

【摘要】：交通系统是一个包含人、车、路的复杂系统，在整体上具有宏观特性的同时，其内部各个要素之间又有着不同的个体差异和一定的随机性，这些特性与元胞自动机自身具有的特点十分相近。为此，本文将元胞自动机理论引入到交通流领域，通过对微观交通流的建模与仿真，以揭示交通流内在特性与演化机理。本文在Nagel-Schreckenberg一维单车道元胞自动机模型（简称NS模型）和交通流流量、速度、密度关系的基础上，建立了一种考虑单个车辆速度受交通流密度影响的一维元胞自动机单车道微观交通流模型。该模型根据单车道交通流密度的实时变化，能够自适应地判断道路处于自由流状态、亚稳态、还是处于拥堵状态。对所有速度为零的车辆引入了慢启动规则，针对自由流状态和其它状态（包括亚稳态、拥堵状态）提出了不同的速度更新规则。在自由流状态时，采用自由流速度更新规则，车辆受随机慢化影响较小，通过估计前车速度，解决了NS模型中模拟车速小于实际车速的问题。在其它状态时，，采用保守的速度更新规则，通过改进随机慢化规则，降低了车辆速度。该模型有利于提升模拟交通量，可以更加真实的体现单车道交通流特性。通过MATLAB编程实现单车道微观交通流的时空图模拟，结果表明，该模型能够较好的再现单车道自由流、亚稳态、拥堵扩散等现象，与NS模型、近似考虑前车速度效应的模型（简称VE模型）相比，该模型在自由流状态和其它状态的平均交通流速度以及交通流量有显著的提升。该模型的建立对研究单车道交通流特性具有重要意义，有利于进一步研究交通拥堵问题。
[Abstract]:The transportation system is a complex system that includes people, cars and roads. It has macroscopic characteristics as a whole, and at the same time, there are different individual differences and certain randomness among the various elements of the transportation system. These characteristics are very similar to those of cellular automata. Therefore, the cellular automata theory is introduced into the traffic flow field, and the intrinsic characteristics and evolution mechanism of the traffic flow are revealed by modeling and simulation of the microscopic traffic flow. This paper is based on the Nagel-Schreckenberg one-dimensional cellular automaton model (NS model) and the relationship among traffic flow, speed and density. A one-dimensional cellular automaton single-lane microscopic traffic flow model considering the influence of single vehicle speed on traffic flow density is established. According to the real time change of single lane traffic density, the model can adaptively determine whether the road is in a free flow state, a metastable state, or a congested state. The slow start rule is introduced for all vehicles with zero speed, and different speed update rules are proposed for free flow state and other states (including metastable state and congestion state). In the free flow condition, the free flow velocity updating rule is adopted, and the vehicle is less affected by the random slowing down. By estimating the speed of the vehicle in front of the vehicle, the problem that the simulated speed in NS model is less than the actual speed is solved. In other states, conservative speed updating rules are adopted, and the vehicle speed is reduced by improving the stochastic moderating rule. The model is helpful to enhance the simulated traffic volume, and can more truly reflect the characteristics of single lane traffic flow. The spatiotemporal simulation of single-lane microscopic traffic flow is realized by MATLAB programming. The results show that the model can reproduce the phenomena of single-lane free flow, metastable state, congestion diffusion and so on, compared with NS model. Compared with the VE model, the average flow velocity and traffic flow rate of the free flow state and other states are significantly improved. The establishment of the model is of great significance to the study of the flow characteristics of single lane traffic and is beneficial to the further study of traffic congestion.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP301.1;U491.112

【参考文献】