Research On The Impact Of Torrential Rain On Traffic Flow Characteristics Of Expressways

With the development of road networks in megacities, expressways have become the urban traffic artery. Accompanied by the increase in the number of motor vehicles, traffic congestion and traffic accidents occurred frequently. To alleviate these problems, the traffic flow characteristics of expressways become a hotspot for researchers. However, the previous studies are all based on dry weather. When confronted with adverse weather, the behavior of drivers, the traffic operation and traffic flow characteristics would have some changes, and the safety of traffic system would at a risk. Although many studies have shown that rainfall has some impact on the road traffic operation, the torrential rain is not taken into account. There is still a vancancy in studying of impact of torrential rain on traffic flow characteristics, resulting in the lack of theorical basis and guidance in making traffic emergency management strategies, even making the expressway traffic system easily paralyze under torrential rain conditions. Therefore, it is imperative to study the impact of torrential rain on traffic flow characteristics of urban expressways.On the basis of the real traffic data on third ring road expressway in Beijing, this paper compared traffic flow characteristics between torrential rain and dry weather through a multilevel approach. This paper mainly works as follows:First, the torrential rain impact is assessed at a microscopic level:characteristics of lane use and distribution of spacing, time headways and speed. The results show that the differentiation of percentage of lane use is less obvious under torrential rain conditions. As with previous studies, the speed reduces and the time headway increases. But what is different with previous studies is that spacing decreases under torrential rain conditions. In addition, the continuous distribution is more suitable for describing the time headways distribution.Further, the same data were used to extend to a macroscopic level. Four essential parameters used to characterize the Van-Aerde speed-flow-density relationship are calibrated to compare the difference of fundamental diagram between two weather. Results indicate that traffic flow characteristics has some changes under torrential rain. In details, free-speed and road capacity decreases that are consistent with the previous studies. But what made it different is that under torrential rain speed at capacity and jam density increases. Moreover, the curve of fundamental diagrams have some different characteristics under torrential rain compared to dry weather. At last, in order to compare and analyze the applicability of macroscopic traffic flow model to torrential rain, choosing three classical macroscopic traffic flow models, this study used same data to simulate the traffic flow under two weather and use genetic algorithm to calibrate three traffic flow models. Calibration results show that the models applied to describe the traffic flow characteristics under dry weather are not suitable for torrential rain conditions.