A Study of Calibration Parameters in Mixed Traffic Vehicle Propagation Model

• Main Authors: Chi-Hui Wu, 吳契輝
• Other Authors: Tien-Pen Hsu
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/45115771004075495791

碩士 === 國立臺灣大學 === 土木工程學研究所 === 100 === The vehicular flows in Taiwan are mixed with trucks, buses, passenger cars and motorcycles, for example motorcycles ratio is higher than 60% in the cities, it is established in the traditional mixed traffic model, use of motorcycle traffic is transferring by equivalent number of passenger cars, the physical size of motorcycles is smaller than car, and it can easily pass over passenger cars. For these reasons, traditional traffic flow model can’t reasonably explain the mixed traffic flow. Early Greenshields’s model is linear relationship of speed and density, until recent years, empirical results indicate that the speed and density diagram for the triangular shape, linear model is unable to describe the speed of all traffic status, in this study use DynaMIT dynamic traffic assignment model response to actual traffic relationship of speed and density. In this study uses actual survey video to analyze the traffic behavior movement at both inner lanes and mixed lanes, the results can be displayed inner lane free speed will be greater than the mean free speed of the mixed lane, in this model respectively establish road propagation model of inner lane and mixed lane, with the level of the mixing ratio corresponds to a different mix of lane free speed. In this study use area ratio and density correction factor to build sections of the mixed traffic flow propagation model, through the concept of dynamic vehicle length and dynamic vehicle width estimate minimum density and congestion density cut-off point, and use actual survey to estimate the predictive value of mixed traffic sections to road propagation model parameters, the results can show the concept of using area ratio to estimate the advance rate of the mixed traffic flow is feasible.