基于FCM的无检测器交叉口短时交通流量预测  

Short-term Traffic Flow Prediction of Non-detector Intersections Based on FCM

作　　者：张明辉[1] 

机构地区：[1]长安大学信息工程学院,陕西西安710064

出　　处：《计算机技术与发展》2017年第4期39-41,45,共4页Computer Technology and Development

基　　金：国家自然科学基金资助项目(50978030);中国博士后科学基金(2012M521729)

摘　　要：随着城市中的交通路网规模越来越大,要想达到实时、准确的短时交通流量预测目标,其中城市交通的动态诱导,是解决城市交通拥堵的一个重要手段。准确的短时交通流量预测是动态交通正确诱导的基础,尤其是无检测器交叉口的流量预测。但是由于成本等问题,交通流量检测设备并不能覆盖所有交叉口,考虑到城市交通流量的高度复杂性,常规的方法难以对其进行准确预测。模糊C均值聚类分析方法是一种模糊数据挖掘方法,使用该方法对城市路网中各个交叉口进行模糊聚类,获得它们的聚类模式,由于同一模式的样本具有高度相似性,可以使用同一模式下的有检测器交叉口的交通流量预测无检测器交叉口的交通流量。实验结果表明,该方法容易实现且具有较好的预测精度。As the urban road network is larger and larger,in order to achieve the goal of real-time and accurate short-time traffic flow prediction, the dynamic induction of urban traffic is an important means to solve urban traffic congestion. Accurate short term traffic flow forecasting is the basis of dynamic traffic induction, especially that of the non-detector intersectioti. However, due to the cost and other is- sues, the traffic flow detector equipment does not cover all the intersection, and considering the high complexity of urban traffic flow, the conventional method is difficult to predict. Fuzzy C-means clustering analysis is a kind of fuzzy data mining method, using the freeway network in each intersection of fuzzy clustering, to get their clustering patterns. Because the same pattern sample has high similarity, the same mode of the traffic flow of detector intersection has been used to forecast the traffic flow of non-detector intersection. The experi- mental results show that the proposed method is easy to implement and has high prediction accuracy.

关 键 词：短时交通流量 预测 模糊C均值聚类 检测器 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]