An Intelligent Control Model Of Freeway Traffic Flow Under Unfavorable Conditions

With the development of society and economy, the construction of freeway has entered a stage of rapid development, gradually forming the network of "Five Vertical and Seven Horizontal". Due to the increasing number of vehicles, freeway accidents have been increasing year by year. Additionally, with the impact of a variety of adverse conditions, a higher standard for operation and management of freeway is required.According to the actual condition of driving safety on freeway, this paper will classify adverse driving conditions, including:adverse weather factor (rain, snow, fog, strong wind, etc), static bottleneck sections (tunnels, bridges, long downhill sections), and dynamic bottleneck sections (large car,"ghost jam", traffic accidents and vehicle breakdowns). The roads are divided into sections according to the unfavorable factors. Meanwhile, on the basis of division of freeway management units, we setup monitoring system and other information acquisition systems, in order to maximize the usefulness of freeway.Within each management unit, traffic information acquisition devices provide relevant data necessary for traffic control. It also builds dynamic control unit base on traffic flow characteristics, road operation conditions, weather conditions, and other relevant information, in order to management traffic flow of each unit by section. We also introduce the concept of control density, and use control density to build the automatic control model of freeway traffic flow under adverse conditions, in order to improve road capacity, relief traffic congestion, and smooth traffic flow.In this paper, Hebei Provence section of Freeway G2is used as the object of research, and VISSIM simulation software is used to analyze with real example. We conduct simulation analysis according to real traffic flow of workday, holiday, weather condition under peak-hour and valley-hour. Using the travel time of controlled and uncontrolled road section as the control data, the result shows that controlled method is better than uncontrolled method. The more traffic volume, the more obvious the effect.