Research On Key Technologies Of Highway Traffic Control Under Abnormal Events

At this stage,the development model of China’s highways has shifted from "emphasize construction" to "neglecting management",and the core goal of management is to improve traffic efficiency while ensuring the safety of highway traffic.However,the number of highway traffic accidents is still high,the accident itself and induced secondary accidents not only cause a large number of casualties and property damage,while causing widespread traffic congestion,seriously affecting the efficiency of traffic.The highway is a complex system composed of people,vehicles,roads,environment,etc.The abnormal components of the system are the direct cause of traffic accidents.Among them,accidents caused by human errors and poor vehicle technical conditions are sporadic,only through technical means to quickly obtain information on the occurrence of accidents,and use reasonable control methods for rescue and guidance to reduce the consequences of accidents;Accidents caused by meteorological environmental factors have regularity,and corresponding control measures can be taken for different adverse meteorological conditions to ensure driving safety without road closures.With the development of Internet of Things technology and its application in the transportation field,the real-time acquisition of highway traffic flow operation status parameters and meteorological environment parameters has been achieved.Based on this,the thesis collectively refers to traffic accidents and adverse meteorological conditions as abnormal events on highways,and focuses on the following contents:1)A traffic accident automatic detection algorithm is proposed to address the problem of difficulty in obtaining comprehensive,timely,accurate,and direct information about highway traffic accidents.The algorithm takes the variance of speed,occupancy,and flow before and after the event,as well as upstream and downstream,as the eigenvalues.Rough set theory is used to reduce the data,and the reduced attribute set is used as input to the neural network algorithm.The output is used to determine whether there is a traffic accident;2)To address the problem that the current highway traffic accident diversion control method ignores drivers’ willingness,a method of determining drivers’ path selection behavior based on Logit model is proposed on the basis of driver SP survey,and the average driver induced obedience rate is calculated to verify that the highway traffic flow induced method under accident conditions can simultaneously improve individual passage efficiency and accident congestion dissipation speed;3)In view of the current situation that the safety control standard of all-weather highway passage under severe weather environment is not uniform,based on the historical traffic accident data of highway in Shandong Province,analyze the accident occurrence law under different severe weather conditions,take foggy day as an example,based on the driving simulation test data statistics derived from the foggy day highway human-vehicle-road microparameter relationship model,calculate the variable speed limit control strategy,rely on the relationship between the driver’s visual distance and vehicle braking distance in foggy day to calculate the model limit control strategy,select the appropriate index to assess the vehicle driving risk according to the traffic flow theory,and then calculate the flow limit control strategy.The comprehensive research mentioned above has resulted in the formation of an early warning technology system including automatic identification of highway traffic accidents,traffic flow induction after accidents,and safety control of traffic flow under severe weather conditions.