复杂路网环境下高速公路二次事故预警研究

发布时间：2018-02-08 09:48

本文关键词： 二次预警指标体系警限区间神经网络信息发布　出处：《长安大学》2014年硕士论文　论文类型：学位论文

【摘要】：高速公路的兴建是交通事业向现代化迈进的标志，，它带给了人们巨大的经济效益和社会效益,但同时也给其管理者和使用者带来了不可弥补的重大财产损失和人员伤亡。随着经济的不断发展，人们的生活水平不断提高，在快速发展的道路形式多样化的基础上，其出行方式也开始呈现出多元化，人们的出行环境日渐复杂。因此，一旦有交通事故发生在复杂路网环境下的高速公路上,就会极易产生重、特大交通事故，由此而导致的“二次事故”便成为了高速公路的一大杀手，高速公路二次事故的预防成为高速公路安全运营的关键。为了避免二次事故的发生，一大任务也就落在了高速公路运营管理者的身上：科学合理的分析路网环境下二次事故发生的影响因素，及时做好交通灾害预测以及高速公路二次事故的预防。本文的研究思路是按照预警理论“明确警情—寻找警源—分析警兆—预报警度”，搜集资料并对比分析，总结国内外关于路网环境下高速公路二次事故预警的研究现状，以此为基础，首先从高速公路二次事故预警的理论基础、理论框架以及预警系统的构建三个方面出发对预警体系的基础理论进行介绍，并针对交通流参数分析一次事故对交通的时空影响以及对路网的影响，剖析高速公路在常态条件和事故条件下的事故成因；其次，基于构建指标体系的原则，构建二次事故预警指标体系，并对各预警指标进行测算；再次，对指标值的分布形式进行分析，采用正态分布的3σ原理、对比分析以及专家调查等方法对各预警指标的五个警限区间进行设置，给出神经网络方法下五个等级的对应输出向量；最后选择BP神经网络方法作为二次事故预警系统评价方法，建立二次预警的BP神经网络预警模型，并研究预警信息的发布方法和复杂环境下的信息发布内容，最后对不同的预警等级以及特殊重大交通事故分别研究管理策略。
[Abstract]:The construction of the expressway is the symbol of the modernization of the transportation industry. It has brought great economic and social benefits to people. But at the same time, it has also brought irreparable heavy property losses and casualties to its managers and users. With the continuous development of the economy, people's living standards are constantly improving, and on the basis of the diversification of the road forms of rapid development, Its travel modes are also beginning to be diversified, and people's travel environment is becoming more and more complex. Therefore, once there are traffic accidents on expressways under the complex road network environment, it is easy to produce heavy and serious traffic accidents. As a result, the "secondary accident" has become a major killer of the expressway, and the prevention of the secondary accident of the expressway has become the key to the safe operation of the highway. A major task also falls on the expressway operation manager: scientific and reasonable analysis of the impact factors of secondary accidents under the road network environment, timely traffic disaster prediction and highway secondary accident prevention. The research idea of this paper is to collect and compare the data according to the theory of early warning "clear alarm situation-search for warning source-analysis warning sign-warning degree", and summarize the current research situation of expressway secondary accident early warning under road network environment at home and abroad. Based on this, first of all, the basic theory of the early warning system is introduced from three aspects: the theoretical basis, the theoretical framework and the construction of the early-warning system. According to the traffic flow parameters analysis of a traffic accident on the traffic space and time impact and the impact on the road network, analyze the highway accident under the normal and accident conditions. Secondly, based on the principles of building an index system, The index system of secondary accident early warning is constructed, and each early warning index is measured. Thirdly, the distribution form of index value is analyzed, and the 3 蟽 principle of normal distribution is adopted. The corresponding output vectors of the five levels under the neural network method are given by setting the five warning limits of each early warning index by means of comparative analysis and expert investigation. Finally, the BP neural network method is selected as the evaluation method of the secondary accident early warning system, the BP neural network early warning model of the secondary early warning is established, and the method of issuing the early warning information and the content of the information release under the complex environment are studied. Finally, management strategies are studied for different warning levels and special major traffic accidents.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U492.8

【参考文献】