基于流体逼近的路段交通流动态加载研究

路段交通流动态加载是动态交通网络分配模型中的重要组成部分,在路段交通流动态加载中,离散形式或者连续形式的点排队模型被广泛使用.本文在以往研究的基础上,提出基于流体逼近的连续型点排队模型,克服了原有点排队模型中排队负值的情况.通过分析可以发现,原有点排队模型实质上属于具有单一服务台和无限容量的排队模型,基于流体逼近的思想重新定义了原有点排队模型.其中3个主要部分是流量守恒模型,车辆流出模型和时间相关的服务台模型,这3个模型全部都是连续的.由于连续点排队对计算需求较高,本文将连续的排队模型离散化,模拟了3种不同场景下车辆驶出、路段排队情况.本文模型克服了原有点排队中的负排队现象,并且排队过程满足先进先出的原则,模型具有良好的模拟效果.

Dynamic Link Traffic Flow Loading Based on Fluid Approximation

Dynamic link traffic flow loading is an essential component in the dynamic traffic assignment. In the loading models, point-queue model is widely used in discrete-time or continuous-time form. Based on the previous work, this paper briefly reviews the continuous time point queue based on the fluid approximation, which overcomes the negativity problem of original queue. Based on the observation that the original pointqueue model is actually a queuing model with a server and infinite buffer, a fluid flow approximation model is proposed to interpret the original point-queue model. Three essential components are a flow balance function, an exit flow function and a time-dependent capacity utilization ratio function, which are all in continuous form. Finally, with the discretization of the proposed model, outflow and queue are simulated on the link in three different scenarios, and the non-negativity of queue and flow first in first out (FIFO) can be easily observed.