【作者】 殷俊杰；

【导师】 罗霞；

【作者基本信息】 西南交通大学 ， 交通运输规划与管理， 2013， 硕士

【摘要】 随着社会经济的快速增长,城市交通拥堵问题也越来越突出。合理分析道路的交通状态,是交通控制系统与交通流诱导系统协同的重要依据。有效地对当前路段交通状态进行识别,并及时地预测下一时刻交通状态,可以为交通参与者提供出行诱导,从而避开交通拥挤路段,减少出行时间,进而缓解道路拥堵；交通管理者也可以实时地掌握道路的整体运行情况,便于做出正确的交通决策。现在很多城市的道路上安装了交通检测设备,采集了多种交通数据,如何利用大量的异构数据进行交通状态判别及预测具有一定的研究价值。快速路是城市路网的主要组成部分,承担城市中大部分交通出行量,城市路网的交通状态与出行质量在很大程度上可以从快速路的交通状态上体现出来。本文利用实际的微波传感器数据,研究了快速路交通状态识别及预测方法。主要的研究工作如下：首先对城市快速路交通流数据的采集与预处理进行了研究,分析了速路交通流参数的特征参数,为交通状态的识别和预测提供了理论支持和数据基础。其次,在详细分析和总结前人对交通状态相关研究成果的基础上,界定了交通状态概念,选取了交通状态判别指标。针对快速路交通状态数据特点,应用改进的模糊聚类方法对交通状态判别研究,分析了各交通参数在聚类中的特征权重,确定了交通状态划分的最佳类数,并用实际数据对本研究提出的判别方法进行了误判验证。最后论文将Probit理论引入到交通状态预测,构建了基于Ordered Probit回归的快速路交通状态预测模型,直接对快速路交通状态进行预测。为了验证模型的有效性和准确性,本文采用了统计检验、模型预测与实测数据对比两种检验方法,验证过程表明模型预测结果比较准确,可信度较高。更多还原

【Abstract】 With the rapid growth of the social economy, the problem of urban traffic congestion has become increasingly acute. Reasonable analysis of the state of the road traffic plays an important role in coordination of traffic control systems and traffic flow guidance system. The identification of road traffic state, and predict the next time traffic status in a timely manner could induce travel traffic participants, so as to avoid traffic congestion sections, reducing travel time, thus alleviating road congestion; On the other hand, traffic managers can also real-time grasp the overall operation of the road, easy to make the right transportation decisions. Now traffic detection equipment have been installed on many city roads, gathering a wide range of traffic data, and how to make use of a large number of heterogeneous data to identify and forecast traffic state has certain research value.The expressway is a major part of the urban road network, undertaking the most amount of traffic travel in the city, and to a large extent the freeway traffic state reflects the traffic state of the urban road network and travel quality. In this paper, by using of a microwave sensor data, I make some research on freeway traffic state identification and forecasting methods. The main work is as followsFirstly analysis the way of acquisition and pre-processing on urban freeway traffic flow data,and elaborate the characteristic of traffic flow, thus provides theoretical support and data base for the identification and prediction of the traffic state.Secondly, on the basis of detailed analysis and summary of previous research on traffic state, the defined the concept of the traffic state, selected the discriminant index of state identification, used the improved fuzzy clustering method on traffic state discrimination, analyzed the transport parameters in the clustering feature weight, determine optimal class number in the traffic status clustering. Discrimination method was proposed in this study by using the actual data to test the improved method.Last Probit theory is used in the traffic state prediction, freeway traffic state prediction model was constructed based on the Ordered Probit regression. In order to verify the validity and accuracy of the model, statistical tests and model predictions to compare was used, and the results show that the model is accurate and the credibility is high.更多还原