高速公路交通运行状态判别方法研究

发布时间：2019-02-13 06:51

【摘要】：在经济迅猛增长的今天，城市交通运行态势日趋严峻的同时高速公路从原来的解决制约经济运输发展的助力，已经开始演变成又一个亟需缓解的问题。伴随着我国的汽车保有量逐年大幅度的上升，我国的许多高速公路上也相继产生了交通拥堵的问题并有加剧的趋势，整个高速公路路网急速驶入“拥堵时代”。针对以上现象，本文从交通运行状态含义而提出高速公路运行状态体系，从基于对于高速公路的环境来具体分析高速公路运行状态体系的需求出发，进而建立运行状态体系。本文从设计原则到指标选取，从体系的搭建到状态指标的量化进行了详细的论述。在对交通状态指标体系架构分析的基础上，本文确立了由区间交通参数决定的交通拥挤评价标准和由地点交通参数决定的交通拥挤评价标准，针对高速公路中这个特殊交通运行环境提出交通运行状态判别的算法。算法结合交通拥挤评价标准的要求，评判模型由地点参数决定的判别、由区间参数决定的判别和由数据融合决定的判别联合构成。研究分别对各方法进行了细致的阐述，最终达到可靠分析交通状态，，合理监控、分析并实施预警，实现交通管理服务的目的。同时本文分析了交通紧急事件检测的重要作用，从高速公路交通事件检测的基本原理剖析，论述了高速公路交通事件的影响因素，概述了自动检测交通事件经典算法，并以此提出了基于支持向量机的高速公路交通紧急事件判别算法，接着对支持向量机参数进行最优值寻找，又引入改进的粒子群算法，提出了基于改进的粒子群-支持向量机的高速公路交通紧急事件判别模型。通过基于网络搜索的参数寻优实验、基于基本粒子群的参数寻优实验和基于改进粒子群的参数寻优实验，验证了本文设计算法的性能。
[Abstract]:With the rapid growth of economy, the situation of urban traffic operation is becoming more and more severe. Meanwhile, the expressway has become another problem that needs to be alleviated from the original solution to restrict the development of economic transportation. With the increase of the number of cars in our country, the traffic congestion on many expressways in our country has also been increased, and the whole highway network has rapidly entered the "congestion era". In view of the above phenomenon, this paper puts forward the expressway running state system from the meaning of the traffic running state, and starts from the concrete analysis of the expressway running state system based on the environment of the expressway, and then establishes the running state system. This paper discusses in detail from design principle to index selection, from system construction to quantification of state index. Based on the analysis of traffic state index system, this paper establishes the evaluation criteria of traffic congestion determined by interval traffic parameters and traffic congestion evaluation standards determined by location traffic parameters. This paper presents an algorithm for judging the traffic running state in this special traffic environment in expressway. The algorithm combines with the requirements of traffic congestion evaluation standard, and the judgment model is composed of location parameters, interval parameters and data fusion. The methods are elaborated in detail, and the traffic condition is analyzed reliably, the traffic condition is monitored reasonably, the early warning is analyzed and implemented, and the purpose of traffic management and service is realized. At the same time, this paper analyzes the important role of traffic emergency detection, analyzes the basic principle of expressway traffic incident detection, discusses the influencing factors of expressway traffic incident, and summarizes the classical algorithm of automatic traffic event detection. Based on support vector machine (SVM), a highway traffic emergency identification algorithm is proposed, and then the optimal value of support vector machine parameters is found, and an improved particle swarm optimization algorithm is introduced. An improved particle swarm optimization (PSO)-support vector machine (SVM) based expressway traffic emergency discrimination model is proposed. The performance of the proposed algorithm is verified by the experiments of parameter optimization based on network search, parameter optimization based on basic particle swarm optimization and parameter optimization based on improved particle swarm optimization.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491

【参考文献】