Dynamic and static fusion partition method of road units fortraffic congestion analysis(PDF)
长安大学学报（自然科学版）[ISSN:1006-6977/CN:61-1281/TN]
- Issue:
- 2022年2期
- Page:
- 103-114
- Research Field:
- 交通工程
- Publishing date:
Info
- Title:
- Dynamic and static fusion partition method of road units fortraffic congestion analysis
- Author(s):
- PEI Yulong; LIU Rui; CAI Xiaoxi
- School of Traffic and Transportation, Northeast Forestry University, Harbin 150040, Heilongjiang， China
- Keywords:
- traffic engineering; road unit division; cluster analysis; traffic jam; dynamic data
- PACS:
- -
- DOI:
- -
- Abstract:
- In order to solve the defect of weak connection between the existing urban road unit division method and the characteristics of frequent traffic congestion, a road unit division method which can integrate road static attributes and dynamic congestion data for urban traffic congestion analysis was proposed. Firstly, the road was divided into several static road units based on the road nodes such as section capacity nodes, intersection nodes, interchange nodes and entrance and exit nodes in the road static attributes. Secondly, congestion data and Kmeans clustering method, which was used to divide the road into multiple dynamic road units including congestion trend section and unblocked trend section. Thirdly, according to the traffic flow operation law, the fusion division method of static road unit and dynamic road unit under the relationship of inclusion and inclusion was described, and the final road unit was generated through the fusion division method, and it was endowed with similar road static attributes and dynamic congestion characteristics at the same time, which can distinguish the congestion characteristics between different road units in the same urban road section. Finally, the line from Kangan road to Jinxiang street in Harbin was taken as the experimental road for empirical analysis. The results show that after the urban road with a length of 10.38 km is divided by this division method, the twoway road is divided into 24 sections and 20 sections respectively, compared with the traditional indefinite length division method, the division precision is increased by 41% and 18% respectively, indicating that this road unit division method can make a more detailed division of longdistance urban roads. The road conditions at 17:30, 18:00 and 18:30 on the 15th, 16th and 59th days of the case are analyzed with the road congestion tendency obtained after division, the matching rates are 84.6%, 77.4% and 93.8% on the 15th day, 85.1%, 88.9% and 78.8% on the 16th day and 78.4%, 73.1% and 75.0% on the 59th day, the matching rate in the 15th and 16th days are high, which proves that the accuracy of the results of this division method is greatly related to the timeliness of congestion data. 4 tabs, 13 figs, 32 refs.
Last Update: 2022-04-27