基于GA模糊聚类的城市快速路交通流状态划分

Phase Classification of Urban Expressway Traffic Flows with GA-based Fuzzy Clustering

采用基于遗传算法的模糊聚类划分城市快速路交通流状态,借助模糊相关度函数分析聚类结果的有效性,以寻找交通流状态的最佳分类数。遗传算法的引入,有效改善了传统模糊聚类对初始化敏感以及容易陷入局部最优解的问题。经上海市城市快速路实测数据验证,此算法对交通流状态的划分具有可行性,且聚类有效性分析能够合理确定交通流状态最佳分类数。上述研究成果可用于城市快速路交通信息发布、服务水平评价以及交通设施的控制与管理。

In this paper,Genetic-Algorithms-based Fuzzy Clustering is adopted to identify phases of traffic flows on urban expressways and Fuzzy Relevance Function is applied to evaluate the validity of clustering results,in order to decide the best classification of traffic phases.Comparing with conventional Fuzzy Clustering methods,the adoption of Genetic Algorithm improves efficiently the problem of sensitivity when initializing data as well as avoids getting local optimal solutions.This algorithm,proved by realdata from Shanghai expressways,is efficient to classify traffic flow phases,while the analyses of clustering results can determine the most appropriate number of categories.The researches above can help the release of traffic information,the assessment of service level and the control and management of traffic facilities on expressways.