基于混合遗传算法的城市公交车辆调度计划编制

发布时间：2017-12-28 20:31

本文关键词：基于混合遗传算法的城市公交车辆调度计划编制　出处：《兰州交通大学》2016年硕士论文　论文类型：学位论文

【摘要】：近年来,随着我国经济的快速增长,城市人口数量和城市规模在急剧的增加,同时人们机动化出行频率也在不断的提高,特别是小汽车出行在不断地受到人们的青睐,这就使得城市道路交通愈加拥堵。城市道路交通拥堵问题也已成为影响城市可持续发展的主要因素之一,如何引导人们群众尽可能选择公共交通工具出行?是当前城市交通管理者急需要思考的一个问题。众所周知,我国城市人口密度大,道路资源十分有限,因此大力优先发展公交战略是解决城市道路交通拥堵问题的有效途径和必然选择,从而促进整个城市的可持续发展。公交车辆调度计划是保障城市公交有序运营的前提,一个合理的公交调度方案,既能提高公交公司的效益,又能提高人们选择公交出行的比例;进而有效的缓解城市道路交通拥堵问题。本文所研究的公交车辆调度问题是在时刻表给定的基础上,确定区域运营线路的最小车辆数及各车辆所执行任务的车次链。主要研究内容如下:(1)通过查阅诸多相关学术研究资料,本文阐述了当前国内外车辆调度问题的相关理论、研究方法以及所取得的研究成果。(2)对城市公交车辆调度系统问题进行了系统的分析。本文主要论述城市公交调度的影响因素,诸如城市公交车辆调度的形式及分类、城市公交调度的组织体系、公交时刻表与公交车辆调度的关系等相关理论。(3)城市公交车辆调度问题是整个公交运营单位以及学者研究的重点和难点,可分为单车场车辆调度问题和多车场车辆调度问题。对于多车场车辆调度问题,本文在现有整体法的基础上提出一种解决区域车辆调度问题的新方法:首先新设一个虚拟车场,把区域内各线路要完成的任务车次都当成该新设车场的配车点,而实际车场忽略不计,可以得到最小车辆数以及各车辆的车次链;其次通过预先设计好的启发式算法将各辆车分配到各个车场,从而得到区域车辆的调度计划;最后为了克服遗传算法在求解时容易产生“早熟”、“停滞不前”的现象,本文设计了遗传模拟退火混合算法对模型进行求解,并且对算法进行了详细的阐述。(4)为了验证设计算法的有效性,本文分别从单车场、多车场两个方面验证模型及算法的准确性,通过调节不同参数对算例进行详细分析,从而得到最优的车辆调度计划。这也是对解决大规模、复杂的区域公交车辆调度实际问题进行了一次有益的尝试。
[Abstract]:In recent years, with the rapid growth of China's economy, city population and city size increased dramatically, while the people of motorized travel frequency has been improved, especially the car constantly favored by people, which makes the city more road traffic congestion. Urban road traffic congestion has also become one of the main factors that affect the sustainable development of cities. How to guide people to choose public transportation as much as possible is a problem that urban traffic managers need to think about. As we all know, the density of our urban population is large and the road resources are very limited. Therefore, giving priority to the development of public transport strategy is an effective way and an inevitable choice to solve the problem of urban road traffic congestion, so as to promote the sustainable development of the whole city. Bus dispatching plan is the premise to ensure the orderly operation of urban public transport. A reasonable bus scheduling scheme can not only improve the benefits of public transport companies, but also improve the proportion of people choosing public transportation trips, so as to effectively alleviate the problem of urban road traffic congestion. The vehicle scheduling problem studied in this paper is based on the given timetable, determining the minimum number of vehicles for the regional operation line and the train times chain of each vehicle's execution tasks. The main research contents are as follows: (1) by consulting many related academic research materials, this paper expounds the current theories and research methods of vehicle scheduling at home and abroad, as well as the research results achieved. (2) a systematic analysis of the problem of urban bus dispatching system is carried out. This paper mainly discusses the influencing factors of urban bus dispatching, such as the form and classification of urban public transport vehicle scheduling, the organization system of urban bus dispatching, the relationship between bus schedules and public transport vehicle scheduling. (3) urban public transport vehicle scheduling problem is the focus and difficulty of the whole bus operation unit and scholars. It can be divided into single vehicle scheduling problem and multi depot vehicle scheduling problem. For the multi depot vehicle scheduling problem, based on the existing law on the whole to put forward a new method to solve regional vehicle scheduling problems: first, the establishment of a new virtual depot, the line within the region to complete the task as the train yard of the new car, while the actual car can get ignored. The minimum number of vehicles and the vehicle trips chain; followed by each car will be pre assigned to each station design heuristic algorithm is good, in order to get the regional vehicle scheduling plan; finally, in order to overcome the genetic algorithm in solving prone to "premature" and "stagnation" phenomenon, this paper designs a genetic simulated annealing hybrid algorithm to solve the model, and the algorithm is described in detail. (4) in order to verify the effectiveness of the design algorithm, this paper validates the accuracy of the model and algorithm from two aspects of single yard and multi depot separately, and analyzes the calculation example by adjusting different parameters, so as to get the optimal vehicle scheduling plan. This is also a useful attempt to solve the practical problem of large-scale and complex regional bus dispatching.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U491.17

【参考文献】