客运专线运力资源优化配置研究

发布时间：2018-03-03 03:26

本文选题：客运专线　切入点：运力资源　出处：《北京交通大学》2014年博士论文　论文类型：学位论文

【摘要】：2013年以来,随着宁杭、杭甬、盘营客运专线的相继开通,客运专线新增运营里程1107公里,中国客运专线总里程达到11028公里,“四纵”干线基本成型,客货分线势在必行,这必将极大缓解运能与需求之间的矛盾。主要繁忙干线实现客货分线,“人便其行、货畅其流”的目标将成为现实。这对提高铁路运输能力和安全性、提高运输效率、提高客货运输服务质量都有着重要的意义。但是中国铁路运力资源从总体上看相对短缺的问题在一个很长的时间内依然严重。研究先进的、有效的、适合我国铁路客货分线新模式下的客运专线运力资源优化配置机制,使管理效应倍增,实现铁路运输组织水平、投资回报能力、铁路盈利能力、运输服务质量的显著提高进而最终达到充分利用客运专线运力资源,这是中国铁路建设大发展的同时亟待研究的一个重要的课题。本研究以铁道部项目《铁路运输通道运力资源配置模型研究》为依托,围绕客运专线运力资源优化配置这一研究主题,综述了国内外研究现状,采用了优化理论、神经网络方法、系统建模、定性与定量分析相结合等研究方法,构建了客运专线运力资源优化配置的理论体系。从固定设施、客运专线运输组织相关资源、旅客列车开行方案三方面来研究客运专线运力资源优化配置问题。本文的主要研究内容如下： (1)在客运专线固定设施优化方面,重点研究客运专线客运站层次优化,从客运站数据中提取一些客观的指标数据,利用本文提出的基于自组织神经网络SOM(Self-Organizing Map)和k-means的聚类算法将客运专线中的客运站划分成不同层次。过程是首先使用SOM对客运站数据进行初步分类,得到数据对应的BMU(Best Matching Unit);然后使用DBI(Davies-Bouldin Index)指数确定聚类个数k值；最后使用k-means算法对BMU进行精确分类,得到客运站的层次划分结果。另外,还分别对客运专线线路通过能力和客运专线客运站能力进行研究。 (2)在客运专线运输组织相关资源优化方面,分别对客票定价优化、开行时刻优化和旅客时间花费优化进行研究。以客运专线旅客出行费用最小和铁路企业总收益最大为目标,以客运站始发能力、客运站终到能力、载客能力等为约束建立了客运专线客票定价优化模型。在研究开行时刻优化时,对列车的出发时刻进行计算来判断开行时刻是否合理。在分析旅客旅行时间组成的基础上,以客运专线旅客旅行时间花费最小为目标,以停站次数、线路能力、列车开行数目为约束建立了客运专线旅客旅行时问花费优化模型。 (3)在客运专线旅客列车开行方案优化方面,以前面的研究成果为基础,分别研究开行方案的模型、求解算法和算例。首先,以旅客利益为中心,同时考虑铁路运输企业利益,建立了客运专线列车开行方案的多目标双层规划模型。其次,设计了客运专线列车开行方案优化模型的遗传算法求解过程。最后,以京沪客运专线为例进行仿真实验,得出京沪客运专线旅客列车开行方案。在研究已有文献中有关旅客列车开行方案评价体系的基础上,提出了一种旅客列车开行方案的评价指标体系,包括定性指标和定量指标两类,并应用该指标体系对仿真实验得到的京沪客运专线旅客列车开行方案与现行实际开行方案做了对比分析。
[Abstract]:Since 2013, with the Nanjing Hangzhou, Ningbo, pan Ying passenger line opened, the new passenger dedicated railway operating mileage of 1107 kilometers, China passenger total mileage of 11028 kilometers, "P" trunk basic shape, both traditional and imperative, it will greatly ease the contradiction between transportation and demand. The main busy trunk unloading points, "who has the necessary goal, will become a reality. To improve the railway transport capacity and safety, improve transport efficiency, is of great significance to improve passenger transport service quality.
But China railway transportation resources from the overall perspective of the relative shortage still remains a serious problem in a very long time. The research of advanced, effective, suitable for passenger transport resources optimization allocation mechanism a new mode of China's railway passenger and freight line under the management effect of doubling, realize the organization level of railway transport capacity, investment returns railway, profitability, significantly enhance the transport service quality and to make full use of passenger transport resources, this is the development of China railway construction is an important and urgent research topic.
Based on the project of the Ministry of Railways "of railway transport capacity allocation model based on > around this topic, optimize the allocation of resources and capacity of passenger line, summarizes the research status at home and abroad, using the optimization theory, neural network method, system modeling, the research method of combining qualitative and quantitative analysis, constructed optimization theory the system configuration of the passenger transport resources. From the fixed facilities, the transportation organization resources, passenger train plan three aspects to study on optimal allocation of passenger capacity resources. The main contents are as follows:
(1) in the passenger dedicated line fixed facilities optimization, research station level and optimize the passenger focus, extracting some objective index data from the passenger station data, the proposed based on self-organizing neural network SOM (Self-Organizing Map) clustering algorithm and K-means will be divided into different levels of passenger dedicated railway passenger station. Is the first to use SOM for a preliminary classification of passenger station data, get data corresponding to the BMU (Best Matching Unit); and then use the DBI (Davies-Bouldin Index) to determine the number of clustering index k value; finally use the k-means algorithm to accurately classify BMU, get the passenger station classification results. In addition, each of the passenger dedicated line the carrying capacity and passenger station capacity was studied.
(2) in the transportation organization related resources optimization, respectively on the ticket pricing optimization, time optimization and operation time were studied. The optimization of passenger cost to passenger travel cost minimum and maximum gross income of railway enterprises as the goal, to the passenger station originating passenger station to the end of the ability, ability, capacity constraints such as the establishment of the passenger ticket pricing optimization model. In the research on train schedule optimization, the departure time of the train running time calculation to determine whether it is reasonable. Based on analyzing the composition of passenger travel time, the passenger special line of passenger travel time cost minimum as the goal, with the number of stops, line capacity, train number cost optimization the model established passenger travel restrictions.
(3) in the passenger train operation scheme optimization, based on the previous studies, study the operation scheme of the model, algorithm and examples. First of all, in the interests of tourists as the center, taking the railway transport enterprise interests, and establish a multi-objective bi level programming model of passenger train operation plan. Secondly, the process of genetic algorithm is designed for dedicated passenger traffic lines optimization model. Finally, the Beijing Shanghai passenger dedicated line simulation, the Beijing Shanghai passenger dedicated line passenger train operation scheme evaluation system. Based on the existing literature on the passenger train, a passenger train plan evaluation the index system includes qualitative and quantitative indicators of two categories, the operation scheme and the application of the index system for the simulation of Beijing Shanghai passenger train and now A contrastive analysis of the actual opening plan was made.

【学位授予单位】：北京交通大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：F532;F224

【参考文献】