论文总字数：26074字

摘 要

随着现代交通的快速发展，实时交通信息在道路交通管理与控制中的重要性日益增加。现代交通信息的是通过车辆检测器来收集的，根据检测器的构造和使用的不同的技术，它可以分为感应线圈检测、视频检测、红外线检测、雷达检测、激光检测等，本文研究的就是其中的感应线圈的数据收集及处理方法，环形线圈车辆检测器与其它的检测器相比而言，具有低成本、较高检测精度、可以全天候工作的优点，是目前在实践中应用最为广泛的车辆检测器。

对城市快速路的交通管理越来越得到各级管理部门的重视。为了提高管理水平，不少城市开始在城市快速路上布设感应线圈，以期获得实时交通信息，为快速路交通管理提供数据支持。如何根据感应线圈检测到的数据获取高精度的快速交通状态，是本课题研究的重点。

宏观交通流是由交通结构、速度、流量和密度四个元素构成的流体结构。研究速度、流量和密度这三个元素关系并建立动态的模型是研究交通系统的重要组成部分。多年来，各国学者对三者关系的研究不断深入。1934年Greenshields 提出线性交通流模型，随着拟合方法的不断进步，Greenberg、Underwood 、Edie 等人根据具体交通状况提出了不同的交通流模型。论文总结了前任的研究成果，对三参数关系以及感应线圈的数据处理方法进行了归纳。速度、流量和密度是交通信息三个基本参数，交通工程理论研究是以此三参数相互关系为基础的。目前来说，虽然各界学者对三参数的观点仍有不同，但是参数基本的变化趋势还是保持一致。

中国的研究人员针对中国城市特有的交通特性，通过检测数据分析交通流特性，或者以定性分析为基础研究宏观交通流模型 。自从改革开放以来我国城市建设进入了高速推进的时期，城市内的交通状况也远比国外复杂，因此对于道路的交通状态需要长时间的观察、记录与研究，从而模拟出符合我国国情的交通特性模型。通过目前已有的检测数据来看，国内城市的交通特性与国外有部分相似之处，但是两者的差异也非常明显。同时不同等级的道路交通特性拟合结果也有较大差距，所以我们应该根据当地的检测数据结合已有的理论成果建立相应的动态模型。

关键词：感应线圈、交通流参数、行程速度、动态模型

ABSTRACT

With the rapid development of modern transportation, in road traffic management and control of real-time traffic information demand is increasing. Modern traffic information can be collected by vehicle detector, according to the structure of the detector and the use of different technologies, it can be divided into induction coil detection, video detection, microwave, infrared, radar detection, laser detection, etc., is one of the research in this paper the data collection and processing method of induction line, the annular coil vehicle detector relative to other detector with low cost, the advantages of high precision, all-weather work, is the comparison of widely used vehicle detector.

Management of urban expressway traffic has become more and more received management departments at all levels of attention. In order to improve the management level, many cities began to rapidly the road in the city of induction coil, arranged in order to get real-time traffic information, provide data support for the expressway traffic management. Based on induction coil detected data acquisition precision rapid transit state, is a focus in the study of this subject.

Macroscopic traffic flow is made up of traffic structure, speed, flow and density of the fluid structure of four yuan. Study speed, the mathematical model is set up and the relationship between flow and density is an important part of study the traffic system. Over the years, the research about the relation between the three countries scholars further. 1934 Greenshields linear traffic flow model is put forward, with the development of traffic engineering science, Greenberg, Underwood,

Edie Williams et al., depending on the traffic in different traffic flow model is put forward. In this paper, on the basis of previous studies based on traffic flow measurement information for further analysis and research.

Speed, flow rate and density is traffic information three basic parameters, the relationship between the three model is the basis of the traffic engineering theory research. At present, the understanding of the three parameters for traffic although there are different views, but the change tendency of the parameters of basic consistent.

Domestic scholars on China"s urban traffic characteristic, traffic flow characteristics through the analysis of the testing data or macroscopic traffic flow model based on qualitative analysis research. Urban construction in our country has developed rapidly in recent years, and human traffic motorised abroad is more complex, so the analysis and research the characteristic of road transportation need ongoing, build traffic characteristic model which accords with the situation of our country. Traffic based on the current study found that the urban traffic characteristics compared with foreign although have similarities, but the difference is obviously. At the same time, because the road traffic conditions for the level of road is also different so as to make the model has relative uncertainty. Therefore for traffic characteristics analysis modeling in China need to according to local traffic conditions in combination with the measured data and theoretical research to classify dynamic modeling.

Keywords: induction coil, traffic flow parameters, travel speed, the dynamic model

第一章 绪论 7

1.1 选题背景 7

1.2 选题意义 7

1.3 研究现状 7

1.4 研究内容以及思路 7

第二章 感应线圈工作机理 9

2.1 感应式环形线圈检测器的基本介绍 9

2.3 检测器的组成 11

2.3.1 中央处理器 11

2.3.2 检测卡 12

2.4 检测器主要功能 12

2.4.1 检测器协议的架构 12

2.4.2检测器的算法概述 12

第三章 交通流数据特性与异常数据清理 13

3.1 研究对象的确定 13

3.1.1 研究对象选择 13

3.1.2 研究时间的选择 13

3.2 交通流时变特性分析及参数模型 13

3.2.1 交通量时变特性 14

3.2.2 车速时变特性 14

3.2.3 占有率车速时变特性 14

3.2.4 交通流参数模型 15

3.3 异常数据判断算法 18

3.3.1　逻辑推理算法 18

3.3.2　重复性和连续性算法 18

3.3.3　交通流机理分析算法 18

第四章 感应线圈参数提取及动态估计模型建立 20

4.1 交通流三参数检测法综述 20

4.2 两种可行的速度估测方法 21

4.2.1 对数线性回归模型估算平均车长 21

4.2.2 基于自由流的速度估计算法 22

4.3 动态状态-空间模型 23

4.3.1 基于状态-空间模型的动态估计模型 23

4.3.2 行程车速一阶估计模型 24

4.3.3 行程车速二阶估计模型 25

4.4 卡尔曼滤波的基本原理 25

4.4.1 离散系统的卡尔曼滤波 25

第五章 行程车速估计值质量评价研究 28

剩余内容已隐藏，请支付后下载全文，论文总字数：26074字