Parameter Estimation Of Expressway Exception Events Based On Upstream And Downstream Multisource Data

After the occurrence of the expressway exception events,the accurate and timely estimation of the characteristic parameters of the events,including the occurrence time of the event and the impact on the traffic capacity,is conducive to the traffic management department to further analyze the situation evolution of the traffic flow and control the traffic reasonably after the event.The existing exception events detection algorithms can only judge whether there is an event or not,and can not estimate the occurrence time of the event and the impact on the traffic capacity and other event parameters.Due to the influence of the detection cycle time of the detection equipment and the delay of the detection algorithm,the detection time of many events has a certain delay relative to real occurrence time,which means the detection time is not as same as the occurrence time.Therefore,according to the existing detector distribution and the expressway data sources,it is of great significance to further obtain the parameter estimation of expressway exception events for the improvement of the quality of expressway management.Based on the above,this paper estimates the occurrence time of exception events and the degree of impact on the traffic capacity by combining the vehicle detector data,toll data and manually recorded data.Because the delay between the real occurrence time and the detection time is composed of the time that the event spreads to the detector section and the time that the detection algorithm consumes,firstly,the travel time of the road section based on toll data is estimated,and then the event detection algorithm for the large detection cycle is designed to get the detection time as close as possible to the real occurrence time,finally,the detection time is modified by travel time to complete the estimation of the occurrence.For the influence of exception events on expressway capacity,a scheme combining actual data and simulation analysis is proposed.The main contents of this paper are as follows:(1)Based on the toll data,this paper proposes a method to estimate the travel time of high-speed road section.The actual traffic time of the ramp and the waiting time of entering and leaving the toll station,which are included in the travel time is considered.Using the travel time in the original toll data to estimate does not represent the defect of the main line travel time.With the use of a lot of OD data in toll data,this paper proposes a modified estimation method that positively correlates the length of the road section with the accuracy of the travel time,so as to realize the effective estimation of road travel time,and uses the toll data of some expressway networks in Chongqing to verify the method.(2)This paper proposes an estimation model of the occurrence time of exception events based on the travel time correction.Aiming at the problem that the detection cycle of domestic vehicle detector is much longer than that of foreign vehicle detector,which leads to a large delay when using mature detection algorithm.This paper proposes to use interpolation to preprocess the data and detect the events combining with the event detection algorithm of FCM,so as to obtain the event detection time.Then the travel time estimated by the travel time estimation model is used to calculate the time when the event occurs and affects the detector,and the detection time is modified to complete the actual estimation of the occurrence.The validity of the method is verified by the combination of simulation software and actual data.(3)This paper proposes a model of the influence of exception events on traffic capacity based on detection data.It mainly analyzes the influence of different traffic factors on the traffic capacity through the simulation software.Then it uses the reduction coefficient method,and based on the detection data,the method of combining software simulation and reduction coefficient correction is used to complete the estimation of the impact of the effective traffic capacity of the incident point in the exception events,and the actual events of Chongqing Expressway are used to verify.