【作者】 尹紫红；

【导师】 李远富；

【作者基本信息】 西南交通大学 ， 道路与铁道工程， 2013， 博士

【摘要】 越来越多大跨度桥梁的投入使用为社会经济发展以及交通项目建设做出巨大贡献的同时,迫切需要解决的问题是如何提高其在运营期抵御风险事故的能力。大型复杂多通道交通项目(三桥合建)涉及铁路、公路、市政道路等三个不同行业的项目,其技术复杂、难度极大,此举在国内尚属首例,在国际上尚无先例,甚至技术标准、施工操作规范及控制指标均要单独专题研究；建设管理上存在着立项审批由谁来主管、业主由谁担任、竣工验收由谁来主管、组织等难题。尤其是运营期的风险管理更是没有先例可循。本文以温州瓯江北口多通道合建项目以及西堠门大桥为研究背景,主要研究工作如下：1.基于桥梁运营期风险事故数据库,通过查阅大量文献资料,对比分析常用的风险辨识方法各自的特点,提出结构方程模型对大型复杂多通道交通项目运营期的风险源进行风险辨识,并对主要风险源进行排序,为运营期风险管理奠定基础。2.采用问卷调研的方式对三桥合建项目运营期管理风险进行详细调查。构建三桥合建运营期管理风险评价指标体系,分析现行风险评价方法的优缺点,提出定量分析和定性分析相结合的研究方法对调查所得的问卷数据进行整理分析,利用建立的因子分析法模型以及SPSS分析软件进行统计分析,对三桥合建项目运营期管理风险因子进行分析,找出影响三桥合建项目运营期管理风险的关键因子,为后文的风险控制对策分析提供依据。3.针对影响三桥合建项目运营期管理风险的关键因子提出对策和建议,建立健全利益协调机制和风险应对机制,完善项目运营期的管理风险分担机制以及建立管理风险防控体系,降低项目运营期的管理风险。4.研究大跨度桥梁运营期结构安全风险评价现状,提出基于相关向量机(RVM)的结构安全风险评价体系及流程。对相关向量机回归算法进行介绍,并针对传统相关向量机回归算法的缺点,提出采用遗传算法对相关向量机回归算法进行优化,获得性能优越的相关向量机回归算法。利用遗传算法优化的相关向量机算法对大跨度桥梁运营期结构安全风险关键参数进行预测,以西堠门大桥为研究对象进行试验测试与分析验证。5.提出一种大跨度桥梁运营期结构安全风险改进的灰色理论风险评价方法,基于大跨度桥梁运营期结构安全风险关键参数预测数据,建立大跨度桥梁运营期结构安全风险改进的灰色理论评价模型,以实现对大跨度桥梁运营期结构安全风险的准确评价,并通过实例进行分析计算。6.构建大跨度桥梁运营期结构安全风险评价系统总体结构及功能模型,在此基础上设计大跨度桥梁运营期结构安全风险评价系统数据库,描述大跨度桥梁运营期结构安全风险评价系统界面,最终实现大跨度桥梁运营期结构安全风险评价系统。研究成果可以为类似大型复杂多通道交通项目运营期风险管理提供指导,具有较大的理论价值和实际意义。更多还原

【Abstract】 The application of more and more long-span bridges has made great contribution to the development of social economy and the construction of traffic projects. And at the same time, the problem of prompt solution is how to improve their anti-risk ability in the operation period.Large-scaled, complicated and multi-channel traffic projects (three bridges built together) that relate to three different fileds like railway, highway, municipal road etc, are exceptional cases in China or even unprecedented all over the world for the invovled complex technologies and difficult operations. Also, the technical standards, construction norms and control indexes need to be studied separately; there are some other problems in the construction management such as who will be in executive charge of the project approval, who will be the owner and who will be the supervisor and organizer for the completion and acceptance. Even the risk management in the operation period can follow no precedent.Situated in the research background of multi-channel joint projects in the north of Wenzhou Oujiang and Xihoumen Bridge, this paper aims to do the following research:It is based on the management system of bridge accidents happened in history, comparing and analyzing each commonly used risk identification by consulting large amount of literature. Also, it comes up with a method combining the accident tree method and the analytic hierarchy process to distinguish the risk sources for large complex multimodal transportation project in its operation period. Meanwhile, it sorts out the main risk sources, laying the foundation for risk management in the operation period.It investigates into detail the risk management for Three Bridges Built-together project in its operation period by conducting questionaires. Therefore, it is to build up an evaluation index system of risk management for this project in the operation period, to analyze the advantages and disadvantages of the current risk evaluation method, to put forward a research method combining quantitative and qualitative analysis for the consolidation and analysis of the data obtained from the questionnaires.Also, this research conducts its statistic analysis by using the present Factor Analysis Model and SPSS analysis software to analyze the risk factors of Three Bridges Built-together Project in its operation period. It tries to find out the key factor influencing the risk management in the operation period of this project, and to provide profound basis for the analysis of countermeasures control in the later part.The research aims to put forward countermeasures and suggestions for key factors that may influence the management risk of Three Bridges Built-together Project in its operation period, to establish and perfect the benefit coordination mechanism and the risk-response mechanism, and to improve the risk sharing mechanism and establish the risk prevention and control system for the reduction of management risk in the operation period of the project.The papre conducts researches into the situation of structure safety evaluation for long-span bridges in their operation periods, and puts forward a risk-evaluation system of structure safety and process based on RVM, in which the relevance vector machine regression algorithm is introduced. Meanwhile, it tends to adopt genetic algorithm for optimization of the traditional relevance vector machine regression algorithm, which eliminates the disadvantages of traditional relevance vector machine regression algorithm.To forecast the key parameters of management risk in the operation period of long-span bridges by using the relevance vector machine algorithm optimized by genetic algorithm, the current research takes Xihoumen Bridge as its research object for test and analysisThe paper comes up with a method of grey theory risk evaluation improved in the operation period for long-span bridges, and establishes accordingly a grey theory evaluation model based on the predictive data for key parameters of management risk. The aim is to provide accurate evaluation of the structure safety risk for large-span bridges during the operation through example analysis.The research constructs the overall structure and its function model of the structure safety risk evaluation system for large-span bridges during the operation period, on which base to design its system database and describe the system interface, and ultimately establish the whole system of structure risk evaluation of large-span bridges during its operation.It can provide guidance for risk management of similar complicated multi-channel traffic projects in the operation period, thus embracing great theoretical and practical significance.更多还原