危险品运输网络的可靠性博弈分析及减灾系统研究

发布时间：2018-01-31 08:27

本文关键词： 危险品运输路网可靠性减灾系统博弈　出处：《西南交通大学》2013年博士论文　论文类型：学位论文

【摘要】：由于危险品具有易燃、易爆、毒害等特殊物理和化学特性,一旦承运这些货物的车辆在运输过程中发生泄漏、火灾或爆炸等事故,不仅会给事故沿线居民的生命健康和财产安全造成不可挽回的伤害,还有可能对周围生态环境造成重大损失。因此,如何确保道路危险品运输安全,降低事故风险,对保障人民群众生命财产安全及社会稳定具有重要理论和实际意义。论文首先对路网的可靠性以及危险品运输路网的可靠性进行分析,接着研究危险品运输网络可靠性失效的形成机理,以及影响危险品运输路网可靠性的五个因素：人的不安全行为、危险品理化特性、机械的不安全状态、环境因素、安全管理因素,最后从预防的角度研究运输路径的先期优化原理。研究危险品运输过程中,政府相关的监管部门与危险品运输企业博弈问题。由于信息不对称,在博弈过程中,政府与危险品承运人经过多次博弈最后达到平衡。对危险物品运输网络的设计进行了分析,政府和危险品运输企业的双层约束条件下,在此基础上兼顾政府期望的风险和危险品运输企业期望的成本上不同的利益目标,建立一个双层规划模型,根据非合作博弈论的相关理论设计启发式算法进行求解。危险品运输路网破坏者和危险品运输承运人可靠性博弈模型,斯塔克伯格-纳什模型和部分合作性博弈模型,对他们的联系和功能进行了分析和验证；给出在什么样的情况下不合作博弈可以给出最坏情况下的解决方案；(1)破坏者进行联合后对危险品运输路网出行成本可靠性的影响(2)如果在路网中加一条路线,所有用户的期望出行成本和出行成本的可靠性都变得更糟的悖论现象；接着在通过从投资方面提高危险品运输路网可靠性方面提出了一个二阶段随机规划,第一阶段找出需要投资的危险品运输路段,第二阶段危险品运输路网中O D对之间的最小遍历成本,可靠性成为了目标函数的因子。利用最短路径的算法把原来的问题推导成另外一种等价形式。通过放松对投资变量的整数限制和对泰勒展开式的应用,得到一个多元线性函数,即为投资边际效益问题。最后对危险品发生事故后从人员疏散、车辆疏散、应急物资调度优化方面建立模型,并通过相关算法对模型进行求解。
[Abstract]:Due to the flammable, explosive, toxic and other special physical and chemical characteristics of dangerous goods, once the transport of these goods in the transport of vehicles in the process of leakage, fire or explosion and other accidents. It will not only cause irreparable harm to residents' health and property safety along the accident, but also may cause serious losses to the surrounding ecological environment. Therefore, how to ensure the safety of dangerous goods transportation on roads. It is of great theoretical and practical significance to reduce the risk of accidents to ensure the safety of people's life and property and social stability. Firstly, the paper analyzes the reliability of the road network and the transportation network of dangerous goods, and then studies the forming mechanism of the reliability failure of the transportation network of dangerous goods. And five factors affecting the reliability of dangerous goods transportation network: unsafe behavior of human, physical and chemical characteristics of dangerous goods, unsafe state of machinery, environmental factors, safety management factors. Finally, the advance optimization principle of transportation route is studied from the point of view of prevention. In the process of dangerous goods transportation, the relevant government supervision department and the dangerous goods transportation enterprise game problem. Because of the information asymmetry, in the game process. The government and the carrier of dangerous goods have reached a balance after many games. The design of the transport network of dangerous goods is analyzed under the double constraint conditions of the government and the dangerous goods transportation enterprise. On the basis of this, a bilevel programming model is established, which takes into account the different benefit objectives of the government's expected risk and the expected cost of the dangerous goods transportation enterprises. A heuristic algorithm is designed to solve the problem according to the related theory of non-cooperative game theory. The reliability game model of destroyer and carrier of dangerous goods transportation network, Stackberg-Nash model and partial cooperative game model are analyzed and verified. The solution in the worst case can be given under what kind of situation uncooperative game can give; (1) the impact of saboteurs on the reliability of the travel cost of the dangerous goods transport network after a joint operation) if a route is added to the road network. The paradox that all users expect travel cost and travel cost reliability become worse; Then in the aspect of improving the reliability of dangerous goods transportation network from the aspect of investment, a two-stage stochastic programming is proposed. The first stage is to find out the section of dangerous goods transportation that needs investment. The minimum traversal cost between O-D pairs in the second stage of dangerous goods transport network. Reliability has become the factor of the objective function. The original problem is deduced into another equivalent form by using the shortest path algorithm. By relaxing the integer restriction on investment variables and the application of Taylor expansion. A multivariate linear function is obtained, that is, the problem of marginal benefit of investment. Finally, the model of evacuation from personnel, vehicle evacuation and emergency material scheduling optimization is established after the accident of dangerous goods, and the model is solved by relevant algorithms.
【学位授予单位】：西南交通大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：U492.336.3;U492.8

【参考文献】