基于网格模型与K-Means算法的交通状态演变特征

Evolution characteristics of traffic state based on grid model and K-Means algorithm

为利用海量网约车轨迹数据实现对城市交通状态的高效识别与分类,对成都市网约车轨迹数据进行预处理,构建城市交通状态识别网格模型,根据模型判别网格的交通状态。利用K-Means聚类算法对不同时段的交通状态进行聚类,并将交通状态分为持续畅通型、轻度缓行型、持续缓行型、持续拥堵型4种类型,从时间维度和空间维度分析不同网格的交通状态演变特征。结果表明:研究区域内交通状态相对稳定,持续拥堵区域分散,持续缓行区域较为集中。基于网格模型与K-Means算法的交通状态识别方法能够实现对交通状态的快速判别与聚类,可实现对不同城市交通状态的识别。

In order to realize the efficient identification and classification of urban traffic state by using the massive net-charted cars′trajectory data,the net-charted cars′trajectory data of Chengdu City are pre-processed to establish the grid model of the urban traffic state recognition and judge the traffic state of the grid by the model.The K-Means clustering algorithm is used to collect the traffic state data at different periods,then the traffic states are divided into four types of the smooth flow,mild flow,continuous flow and continuous congestion,and finally the evolution characteristics of different grids of traffic states are analyzed from the time and spatial dimensions.The results show that the traffic states in the researched area are relatively stable,the continuous congestion area is scattered,and the slow flow area is more concentrated.So,the recognition method of the traffic states based on the grid model and K-Means algorithm can implement the rapid measurement and clustering of traffic states and also the recognition of traffic states in different cities.