| 摘要: |
| 为实现能实时、高效、准确地利用现有的车牌识别数据掌握路网交通状态,并精准实施道路交通网络控制措施,基于道路交叉口车牌识别系统数据,依据基于平均最短路径长度的拥堵指标衡量路网运行状态,使用K-means聚类算法进行分类,依据谱距离指标和拥堵指标分类不同时段的网络剩余供给能力。在此基础上,依据实时的道路交通数据,计算拥堵指标,可得到实时道路网络剩余供给能力分级。针对分级后的实时道路网络的实际剩余供给能力,设计不同的网络交通管控方案,能有效缓解交通拥堵。以长沙市五一广场局域道路网络为实例,研究结果表明:城市道路网络状态全天可以分为4种类型,拥堵指标为1.522、1.823、2.207、3.069,分别对应路网剩余供给能力良好、一般、较差和极差。该网络早晚高峰时段路网剩余供给能力多处于较差状态,并且晚高峰时路网有可能处于剩余供给能力极差状态。相比于工作日,在周末的高峰时段,路网剩余供给能力表现较为良好。采取针对性交通管控措施后,经仿真验证表明拥挤路径社会总成本下降5%以上。 |
| 关键词: 城市交通 供给能力 退化路网 K-means聚类算法 谱距离 |
| DOI：10.11918/202202003 |
| 分类号:U491 |
| 文献标识码:A |
| 基金项目:长沙市自然科学基金(kq2208282) |
|
| Classification method of network traffic status based on electronic police data at intersections |
|
LI Maosheng,LI Hangcong
|
|
(School of Traffic and Transportation Engineering, Central South University, Changsha 410075, China)
|
| Abstract: |
| A real-time, efficient, and accurate grasp of road network traffic status is the premise of effective control of road traffic networks. Based on the data of the license plate recognition system at road intersections, this paper measures the road network operation status using the congestion index based on the average shortest path length and then uses the K-means clustering algorithm to classify the remaining supply of the network in different periods according to the spectral distance index and the congestion index. On this basis, according to the real-time road traffic data, the congestion index is calculated, and the real-time remaining supply capacity classification of the road network can be obtained. According to the actual remaining supply capacity of the classified real-time road network, different network traffic control schemes are designed, which can effectively alleviate traffic congestion. Taking the local road network of Changsha Wuyi Square as an example, the research results show that the local road network of Changsha Wuyi Square can be divided into four types throughout the day, and the congestion indicators are 1.2,1.3,2.207, and 3.069, respectively corresponding to good, average, poor and extremely poor remaining supply capacity of the road network. In the morning and evening rush hours, the remaining supply capacity of the network is mostly in a poor state, and the network may be in an extremely poor state in the evening rush hours. Compared with weekdays, the remaining supply capacity of the road network performs well in the rush hours on weekends. After the targeted traffic control measures are taken, the simulation results show that the total social cost of congested paths is reduced by more than 5%. |
| Key words: urban traffic supply capacity degradable transport network K-means clustering algorithm spectral distance |
