基于组合模型的短时交通流的预测研究

发布时间：2018-01-09 08:30

本文关键词：基于组合模型的短时交通流的预测研究　出处：《兰州交通大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着经济的发展和城市化进程的加快，汽车保有量逐年增加，出行者的数量也在急剧增长，这给原本就十分紧张的的交通带来了巨大的压力，交通问题已经成为制约国民经济发展的瓶颈问题，其中，对于人们的生活影响最大的就是交通拥堵问题，智能交通系统是解决交通拥堵问题的有效手段。智能交通系统（Intelligent Transport System，简称ITS）综合应用信息技术、通讯传输技术、电子控制技术和计算机处理技术等，对交通信息进行有效地采集、加工和处理，目的是实时、精确地反映当前及其未来的交通状态，最终实现对交通进行科学的组织和控制。短时交通流预测是ITS的核心内容，本文对短时交通流进行预测旨在得到实时、准确的交通预测信息，指导出行者选择合适的出行路线，节约出行者的旅行时间，达到缓解道路拥堵、减少环境污染、节约能源等目的。本文基于短时交通流的基本理论，针对某断面上的短时交通流量，，依次使用移动平均模型、神经网络模型和组合模型进行了预测研究并运用几种评价指标对比分析了各种模型的预测效果，说明了线性模型的可行性，而且验证了组合模型在短时交通流预测中的优势。本文主要的研究工作如下：（1）概括介绍本文的研究背景和意义，并进行归纳总结，对交通流的特性和可预测性进行分析；（2）介绍短时交通流的数据采集和预处理方法，并简要介绍描述其特性的基本参数；（3）详细介绍移动平均模型、神经网络模型和组合模型的有关知识，并建立了移动平均模型、GRNN神经网络模型和基于移动平均和GRNN的组合模型，并用实测数据进行了验证，得到了满意的结果。
[Abstract]:With the development of economy and city urbanization, increasing car ownership, the number of travelers is increasing sharply, which brings enormous pressure to the already tense traffic, traffic problems have become the bottleneck of restricting the development of the national economy and the people's life, the impact of traffic congestion is the biggest problem, the intelligent transportation system is an effective means to solve the problem of traffic congestion. Intelligent traffic system (Intelligent Transport System, referred to as ITS) the comprehensive application of information technology, communication technology, electronic control technology and computer processing technology, the traffic information collected, processing and handling, to accurately reflect the real time. The current and future traffic state, realize the scientific organization and control of traffic.
Short term traffic flow prediction is the core content of ITS, the short-term traffic flow forecasting is to obtain real-time and accurate traffic information, guide travelers choose appropriate travel routes, saving travel time of travelers, to relieve traffic congestion, reduce environmental pollution, save energy.
In this paper, the basic theory of traffic flow based on traffic flow in a section, followed by the use of moving average model, neural network model and the combination model research and the use of comparative analysis of several evaluation index prediction models, to illustrate the feasibility of the linear model, and verifies the model in traffic flow the prediction of advantages. In this paper, the main research work is as follows:
(1) the background and significance of this study are summarized, and the characteristics and predictability of traffic flow are analyzed.
(2) the data collection and preprocessing methods of short time traffic flow are introduced, and the basic parameters describing the characteristics of the traffic flow are briefly introduced.
(3) introduce the related knowledge of mobile average model, neural network model and combined model in detail, and establish the moving average model, GRNN neural network model and the combined model based on mobile average and GRNN.

【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U491.14

【参考文献】