Traffic Status Identification Based On Clustering Algorithm

Abstract

In order to accurately identify the traffic state of the expressway, this paper preprocesses the ETC monitoring data of the expressway based on the traffic flow parameters of the expressway. A fuzzy C-means clustering model was established to cluster the traffic volume, time average vehicle speed and time occupancy rate data of specific road sections. In order to avoid outliers becoming cluster centers, the data density (DKC) value was used to improve the model. Taking the traffic volume, time average speed and time occupancy rate of a typical section of Suzhou Ring Expressway as an example, clustering calculation is performed to classify the traffic state of this section.