摘要
首先是基于理论上的猜测,依靠统计学对于固定路段的"自由流车速"进行了稳定性分析和验证,得到了固定路段的"自由流车速"是一个稳定值的结论;然后根据这一具有稳定性的交通参数,采用阈值限定和聚类的方法对疑似问题检测线圈进行了排查,并结合常理初步制定了判断规则;为了证实该规则是否有效,有针对性地进行了实验设计;并通过多次不同情况下的实验对判定规则进行了检验和调整;最后对各次的实验结果进行了分析,归纳出了可靠的问题线圈判定依据.实验结果表明,在海量交通数据积存的基础上,利用该方法对可能有问题的检测设备进行甄别具有可行性和科学性.
Theory hypothesis is the premise of this paper. The conclusion that "Free Flow Speed" is a stable value in a fixed section of a highway has been drawn depending on the theory hypothesis and analyzing by making use of statistics, This important result can help us move forward to screening questionable loop detectors which have some problems with them and produce distorted data. Threshold values limitation and clustering approaches are used in this paper to exclude and select questionable loop detectors relying on the stable traffic parameter, And initial criterions to screen questionable loop detectors come into being with the help of these approaches and common practice. Experiments have been designed to approve the validity of these initial criterions. Then several necessary experiments in different situations are successfully carried out to validate and adjust the initial criterions. At last, the reliable determinant criterion is concluded after analyzing the results of the experiments. The results of the experiments show that this method is feasible and proper in screening questionable loop detectors based on the mass archived traffic data.
出处
《交通运输系统工程与信息》
EI
CSCD
2008年第4期30-35,共6页
Journal of Transportation Systems Engineering and Information Technology
基金
上海市科委资助项目(06dz12001)