Intersection passing time period depth division method and system based on space-time state

Abstract

The invention discloses a depth division method and system for intersection passing time periods based on space-time states. The method comprises the following steps: firstly, macroscopically carrying out primary division on a time period through an ordered clustering algorithm based on Fisher optimal segmentation, and then, in order to compensate for the problem that segmentation points are dense in the time period with lower saturation caused by data volatility, carrying out deep division on the time period with a larger time period division range macroscopically; and obtaining a state critical point of the flow in space through a k-means algorithm, and merging segment points in the same state. Through the operation of the three steps, the division condition of the time periods and the flow state corresponding to each time period can be obtained, and corresponding timing schemes are configured for different flow states. Through Vissim simulation and field test verification, compared with fixed timing, the time period division method has the advantage that the queuing length of peak hours and the idle condition of flat peaks are greatly improved.

Description

technical field [0001] The invention belongs to the technical fields of traffic time division, orderly clustering, traffic state identification, intersection signal timing and traffic simulation, in particular to a method and system for deep division of traffic time at intersections based on space-time states. Background technique [0002] In the process of traffic signal control, the commonly used methods include timing control, adaptive control and real-time control. Adaptive control and real-time control have higher requirements for real-time data. At present, there are not many intersections covering radar and real-time detectors in cities, so these two schemes cannot be effectively implemented. Timing control is the most commonly used and easiest to implement signal control method at present. The commonly used timing control methods are divided into two types: fixed time allocation and time-division timing control (TOD). The time-based timing control can more effectiv...

AI Technical Summary

Problems solved by technology

However, due to the lack of consideration of time characteristics, the above algorithm has certain deficiencies in time continuity and accuracy. To solve this problem, ordered clustering (such as the literature "Intersection signal control model based on Algorithm research", "Signal control period division method using recursive orderly clustering", etc.) have been applied to the period division and achieved certain results, but due to the volatility of flow data in different periods, this period division method will be As a result, the segmentation points are concentrated in the time period with low saturation, resulting in the subdivision of the time period is not microscopically detailed, and the results obtained are often not ideal

And the result of the division is only the segmented status of the traffic in the time period, and the traffic status of the time period cannot be identified

Images