The State Evaluation of Highway Traffic Flow Based on Observed Data

摘要

Along with the wide application of the traffic observation and control equipment in the highway, traffic information center can collect a mass of traffic flow data. Analyzing these data can be used to evaluate the highway region's traffic operation state so as to improve traffic control strategy and bring a new research field. Based on comprehensive evaluation of domestic and international traffic flow classification and evaluating indicator selection, this paper put forward a method of K-means clustering algorithm and the EM algorithm to extracting an eigenvector to describe highway traffic flow state of similar region section, and then established the similarity evaluation index system. Finally, take the data of Shaanxi Province Observatory System as an example to verify its actual situation and the conclusion reflects well and has a good effect.