Research On Operating Speed Prediction Of Expressway In Fog

In the development process of expressway,adverse weather condition is a serious threat to traffic safety.Whatever the design concept of expressway,weather conditions were considered to be consistent,ignored the impact of weather differences on the driving environment,especially in mountainous areas where has rich water,and the air is coagulated to form a large area of the fog zone.In this paper,the expressway driving environment and the operation vehicles in foggy environment are put as the main object of study,the different impact degree of road environment on the drivers under different visibility in fog,considered the visibility,curve radius and longitudinal grade as the main factors of affecting operating speed in the fog,established regression prediction model and BP artificial neural network model for small and large vehicles,and then predict the operating speed for small and large vehicles under different constraints.Firstly,according to the analysis of the space-time environmental conditions for the formation of fog and fog classification,the fog influence mechanism on the road environment and the driver's vision are researched.Combining the changes of different factors in fog and the distribution of operating speed,it is identified that the main factors affecting the operating speed are visibility,curve radius and longitudinal grade.Secondly,the mountain expressway is chosen as the main test object,and three expressways near Chongqing are selected as the main test section,the major test data was captured by live and video,and the operating speed and visibility were classified and pretreated.Aimed at vehicle features of the operation vehicles on the expressway,carried out the distribution descriptive statistics of the highway cross-sectional speed for different vehicles,and conducted the correlation test for the operating speed and visibility when the visibility is less than 200m and more than 200m.The result shows that when the visibility is less than 200m,the visibility and the operating speed is highly relevant,when the visibility is more than 200m,visibility and operating speed is less relevant,and when visibility is less than 150m,the road alignment almost has no effect on operating speed.Therefore,the threshold of visibility in fog is set as 200m,and the operating speed prediction model segmentation threshold is 150m.Finally,by analyzing the variation of visibility,when the visibility is less than 150m,the operating speed prediction model which is divided into 50m as a unit was established by using segmented regression method.When the visibility is between 150m and 200m,visibility is the main factor affecting the operating speed;When visibility is between 200m and 500m,curve radius and longitudinal grade are the main factors for the operating speed.Then the prediction model for operating speed was constructed by using artificial neural net work.The predicted result shows that it has good prediction deviation control and error convergence,so the prediction model works well.The research results can provide theoretical basis for improving the expressway safety and ability for smooth flow and the management decision-making for speed in foggy weather,meanwhile it provides theoretical support for the highway route scheme comparison,alignment safety design through fog zone,the traffic engineering facilities and out field facilities setting.