Ramp control method based on congestion situation grading

Abstract

The invention relates to a ramp control method based on congestion situation grading, and the method comprises the steps: carrying out the analysis of historical data, carrying out the clustering through a Gaussian mixture model to redivide a congestion situation, because a critical occupancy is usually difficult to obtain, and the inaccuracy of the occupancy will directly affect the performancesof an ALINEA algorithm. Traffic flow is taken as a control parameter of the ALINEA algorithm, the control rate is adaptively selected according to the congestion situation, a segmentation control method is adopted to increase queuing length constraint to avoid influence of overflow generated by overlong ramp queuing on ground traffic, and the control target is transferred to the ramp when queuingis overlong. The simulation experiment is carried out by utilizing SUMO simulation software, and the result shows that ramp queuing and waiting time of vehicles are optimized to different extents under the condition of ensuring passing of the main line so that the road passing efficiency is improved.

Description

Technical field [0001] The invention relates to the technical field of smart transportation, and in particular to a ramp control method based on the congestion situation classification. Background technique [0002] With the rapid economic development, many cities have built urban expressways to meet the increasing traffic demand, but the number of vehicles has increased year by year, and the problem of traffic congestion has become increasingly serious. At present, the main control methods that can effectively alleviate traffic congestion on urban expressways include ramp control, variable speed limit and traffic guidance. Ramp control is the most widely used and most effective means of expressway traffic control. The traffic flow into the main line is controlled by the signal lights set up on the ramp, so as to achieve the purpose of improving the capacity of the main line. The main methods of on-ramp single-point control are: demand-capacity control, occupancy control, ALINEA...

AI Technical Summary

Problems solved by technology

[0003] Although ALINEA is very effective in controlling single-ramp congestion, including ALINEA and many of its extended algorithms, the control objectives are focused on the control of the main line, ignoring the problem that the ramp is prone to overflow, and the selection of the critical occupancy value of ALINEA has a great influence on the control. The impact is great. If the critical occupancy value is set unreasonably or the occupancy value cannot be obtained, the control effect will be greatly reduced

In terms of congestion situation classification, it is more common to use k-means for clustering, and on this basis to optimize the selection of center points and the number of clusters, reduce the number of algorithm iterations, or increase the accuracy of clustering, but k-means is A hard clustering method that assigns each object to the nearest cluster by calculating the distance between the object and the center point of the cluster. For some cases that are difficult to distinguish, the probability of cluster assignment is lacking, so it is difficult to distinguish

Secondly, the boundary of the k-means cluster is circular, which is not effective when fitting some non-circular data.

Images