交通拥挤事件是城市公共交通系统中造成交通延误的最主要原因之一,快速有效的识别拥挤事件是城市交通控制策略的重要环节。针对交通流相态及其交通因素类属方面存在的模糊性,本文在分析交通流特征时对其进行了聚类软化分。根据交通流特性,运用模糊c均值聚类算法对交通流各要素进行模糊分析处理。通过对交通量隶属度的判别和聚类分析结果,找出不同交通流间的亲疏程度和相似性,将具有相近特性的交通流归纳在一类,从而判别出交通流相态属性,确定交通拥挤事件的发生,达到对交通拥挤事件识别的目的。
Traffic congestion incident is one of the main reasons that cause the traffic delay in the urban transportation system. How to identify these congestion incidents quickly and effectively is an important part of urban traffic control strategy. Traffic status and traffic elements have fuzziness that is hard to categorize. The paper does a fuzzy cluster soft division while analyzing the characteristics of traffic flow. According to the characteristics of traffic flow, it uses fuzzy C-means clustering algorithm to deal with these fuzzy factors. Through the study of the membership degree and the clustering analysis, it gets the degree of closeness and similarity of traffic flow and categorizes the flows. After the comparison and distinguish these characters, it achieves the purpose of traffic congestion incident identification.