【作者】 张春民；

【导师】 李引珍；

【作者基本信息】 兰州交通大学 ， 交通信息工程及控制， 2011， 博士

【摘要】 随着铁路快速发展,路网上编组站需重新布局规划,作为编组站的核心调车设备,驼峰的设计是首先涉及的问题。驼峰设计的合理与否对于提高驼峰解体能力,调车作业安全,减少工程投资,运营费用乃至提高编组站整体作业效率都有着十分重要的作用。驼峰纵断面设计是驼峰设计中的一项主要内容。本文针对目前传统驼峰纵断面设计中存在的问题,以采用“减速器+减速顶”点连式调速系统自动化驼峰为研究对象,在传统驼峰设计理论以及现有研究成果的基础上,运用系统科学,最优化理论,综合评价技术等方法从优化设计方面对驼峰的各组成部分纵断面设计进行了研究。(1)通过分析推送部分的作业特点以及工程和运营要求,从保证驼峰解体效率、作业安全以及减少工程费用出发,建立了推送部分纵断面多目标优化模型。结合模型的特点运用NSGA-Ⅱ多目标遗传算法进行求解,得到了问题的Pareto解集合。(2)分析了溜放部分的作业特点和钩车溜行状态对各坡段设计的要求,以及设计的工程和运营要求,综合考虑能力、效率、费用和安全,建立了溜放部分纵断面多目标优化模型。同样地,运用NSGA-Ⅱ多目标遗传算法进行求解,得到了问题的Pareto解集合。(3)分析了连挂区纵断面不同于溜放部分纵断面的设计特点,根据钩车在该部分溜行状态的要求,考虑效率及投资,建立了连挂区纵断面多目标优化模型。根据模型的特点,提出采用方案比选法对备选的坡度组合方案以及坡长匹配比例关系进行分析计算,确定出了最优坡度组合方案以及最佳坡长匹配比例关系方案。(4)分析了驼峰调车作业中超速连挂、追钩撞车及侧面冲突两类主要调车事故的影响因素。根据驼峰系统的特点以及贝叶斯网络的优点,分别建立了超速连挂事故和追钩撞车及侧面冲突的贝叶斯网络模型。运用贝叶斯网络模型的推理法则求得各影响因素的后验概率,找出影响事故的主要因素。鉴于超速连挂事故是最常发生的调车事故以及目前新型重载车辆的使用,本文进一步对新型重载车辆所引起的超速连挂共因失效问题进行了分析,建立了共因失效混合贝叶斯网络模型,计算分析了新型重载车辆对超速连挂事故的影响程度以及对既有驼峰的适应性。更多还原

【Abstract】 Along with railway’s rapidly developing, the distribution of the marshalling stations on the railway network should be re-planed. As their key shunting equipment, hump design is the critical part firstly involved. It plays an important role in enhancing the hump’s wagon stock disintegrating capacity, ensuring the shunting operation safety, reducing the project investment, cutting the railway operation costs, and promoting the comprehensive operational efficiency.Vertical section design of hump is a main task for the hump design. Therefore, this thesis takes the methods to reasonably design the whole hump vertical section as its research objective on the conditions of the point-continued speed control system with reducers and retarders. To do so, based on the conventional design theory and the existing research results, the thesis takes the deficiency of the conventional design method into account, uses the system science, optimization theory, and comprehensive evaluation techniques.(1) A multi-objective optimizing model for designing the hump pushing section has been established according to the results analyzing the operational characteristics, the construction and operation requirements to the section in order to maintain the higher hump disintegrating efficiency, to ensure the operation safety, to reduce the constructing costs. Furthermore, a Pareto solution set is obtained with NSGA-ⅡMulti-objective genetic algorithm.(2) A multi-objective optimizing model for designing the humping section has been established according to the results analyzing the operational characteristics of the section, the requirements for the car group’s running to the different slope segments of the section, the requirements of the construction and operations, with its humping capacity, operational efficiency, costs and safety taken into account. Also, its Pareto solution set is achieved with the NSGA-ⅡMulti-objective genetic algorithm.(3) A multi-objective optimization model for designing the coupling section has been established according to the results analyzing the different characteristics between the coupling section and the humping section of hump, the requirements for the car group humping status to the section, with the operational efficiency it will have and the construction costs taken into account. Based on the characteristics of the model, the thesis puts forward a scheme comparison method on analyzing and calculating the candidate schemes on grade combination and slope length combination to select the best matched slope length proportion.(4) The effecting factors of two kinds of shunting accidents:over-speed coupling accidents, the rear-end collision and side collision accidents, are analyzed, and the Bayesian network models on the two accidents are established respectively in line with hump-system’s characteristics and the merits of Bayesian network. Then with the inference rule of the Bayesian network model, the posterior probabilities of the main factors are obtained. Since the over-speed coupling is the common accidents, and more and more newly-built heavy-haul cars are put into operation, the thesis also puts one of its research emphases on the over-speed coupling caused by the heavy-loaded cars, and a Bayesian mixture network of common cause failures is set up to calculate and analyze the impact degree of the heavy-haul cars to the over-speed coupling accidents and the adaptability to the existing hump.更多还原