大城市常规公交动态调度理论与方法研究

发布时间：2018-01-19 18:26

本文关键词： 公共交通客流信息动态调度行程时间效益分析　出处：《长安大学》2015年博士论文　论文类型：学位论文

【摘要】：近年来,随着我国综合国力的不断提升,城市公共交通事业得以飞速发展,但与国外先进国家和地区相比,尚存在较大差距,集中表现在信息化、智能化管理与应用水平低下,因此改善城市公共交通环境、实现城市公共交通的动态调度管理,具有重要的理论意义和实用价值。城市公交动态调度管理依托数字化信息采集、处理和网络通信技术,综合运用GIS(地理信息系统)、通信、网络、多媒体及计算机等技术,对城市公共交通系统的信息采集、存储与发布、运营调度及公交优先等进行网络化管理,实现客流变化和车辆运行状况的实时监测和公交资源的优化配置、合理调度。本文探讨了城市公交动态调度管理的结构框架,并就支持系统功能实现的相关技术方法进行了着重探讨,旨在通过对动态管理技术的研究、探索,为我国广大城市提高公共交通管理水平,改善居民出行环境,以解决城市交通拥堵问题,实现城市的可持续发展,提供一定的借鉴依据和参考。本文从城市公交客流的基本特征出发,以西安市的多条线路为分析对象,对大中城市公交客流的时间分布特征、空间分布特征和区域分布特征进行了分类总结,对城市公交线路的增长特征进行了阐述,为城市公交动态调度系统的构建提供了基本理论依据。应用公交IC卡数据对线路客流信息进行了分析、推算,并引入改进的灰色马尔可夫模型来预测线网客流信息,运用最小二乘法和有限差分得到了模型中参数a、b的待定向量,同时引入边值修正因子和马尔可夫精确化来减弱输入数据离散、波动过度的不利影响,提高了线网客流的预测精度。同时,从提高公交服务水平的角度,提出了面向乘客的公交动态信息的发布内容、发布载体和信息传递途径,分析了发布形式和信息更新方式。考虑道路条件、交通负荷、车辆因素、驾驶员等多种影响因素,从路段正常行驶时间、停车时间和延误时间三个角度入手,建立了基于BP神经网络的公交车辆行程时间预测模型。以西安市207路公交车为例,通过调用Matlab软件中的神经网络工具箱,并运用相关调查数据对网络进行训练,并运用检测样本对模型进行检验,将行程时间的预测结果和实际值进行对比分析,分析结果验证了模型的有效性和准确性,为面向乘客的公交动态信息发布系统和公交运营调度管理系统提供了支持技术和理论依据。在对公交调度系统模型和调度形式进行分析介绍的基础上,考虑公交线路站点乘客需求,以乘客候车时间延误最小及车厢内满载率较均匀为优化目标,建立了公交车发车时间表模型,并以车厢满载率为控制参数,对模型进行优化求解,确定高峰时刻的最优发车时间,同时提出了车辆动态调度的分析模型。以西安市13路公交车辆的采集数据对该模型进行有效性检验,分析结果表明应用该模型制定公交车辆发车时刻表,进行车辆运行调度,能够实现线路客流的均匀分布,有效降低乘客的候车时间成本,提高调度效率,并且在线路车辆不足的条件下,运用车辆调度模型进行车辆调度管理,可以快速恢复线路车辆的车头间距,有一定实用价值。最后,针对公交动态管理效益分析的实际需要,提出了直接效益和间接效益的分析指标,具体讨论了减少误乘损失、票价降低、出行时间节约、公交分担率提高、运输费用成本节约及减少交通拥堵所产生的综合效益,并以西安市为例,进行了案例分析,计算结果表明实施数字化信息管理效益显著,展示了良好的应用前景。
[Abstract]:In recent years, with China's comprehensive national strength increasing, city public transportation industry has been developing rapidly, but compared with foreign advanced countries and regions, there are still large gaps, focus on information technology, intelligent management and application level is low, thus improving the city public traffic environment, the dynamic scheduling management of city public transportation, has the important theoretical significance and practical value. The dynamic scheduling management of city bus based on digital information acquisition, processing and network communication technology, the integrated use of GIS (Geographic Information System), communication, network, multimedia and computer technology, information collection of city public transportation system, storage and distribution, operation scheduling and bus priority network etc. management, scheduling optimization, real-time monitoring and public resources to achieve the operation status of passenger flow and vehicle. This paper discusses the city bus dynamic Framework of state management, and were focused on support technology to achieve the function of the system, through the research on the dynamic management technology, exploration, for the majority of city of our country to improve the management level of public transport, improve the residents travel environment, in order to solve the problem of traffic congestion in the City, realize the sustainable development of the city, to provide a certain reference and reference. This article from the basic characteristics of city bus passenger departure, with multiple lines of Xi'an city as the research object, the distribution characteristics of large and medium city transit time distribution and regional spatial distribution characteristics were classified, growth characteristics of city bus lines are described, provides the basic the theoretical basis for constructing the dynamic scheduling system of city bus. Application of IC card data is analyzed, the passenger information calculation, and the introduction of improved grey The Markov model to predict the network traffic information, using the least squares method and finite difference are the parameters in the model a, undetermined vector B, while introducing the boundary value correction factor and Markov accurate input data to reduce the adverse effects of discrete, excessive volatility, improve the prediction accuracy of line network passenger flow. At the same time, from the bus the service level angle, proposed the transit passenger oriented dynamic information publication, publishing carrier and information transmission way, analyzes the release form and information update method. Considering the road conditions, traffic load, vehicle drivers and other factors, influencing factors, from the normal road travel time, three angle parking time and delay time firstly, to establish a prediction model of bus travel time based on BP neural network. The 207 bus in Xi'an city as an example, through the neural network call in Matlab software With the box, and used to train the network related survey data, and using the test samples to test the model, the prediction results of travel time and the actual value of the comparative analysis, the analysis results verify the effectiveness and accuracy of the model, provide technical support and theoretical basis for the dynamic transit channel for the passenger information release system and bus dispatching management system. In the bus scheduling model and scheduling form analysis is presented on the basis of considering the bus stations, passenger demand, passenger waiting time to the minimum delay and the bus vehicle load as the optimization goal, set up a bus timetable model, and car load ratio as control parameter, optimization to solve the model to determine the optimal departure time, peak time, and puts forward the analysis model of dynamic scheduling of vehicles. In 13 Xi'an road public transit vehicle Collecting data for the model validation, the results indicate that the application of the model to develop the bus timetable, vehicle operation scheduling, to achieve uniform distribution of passenger flow, reduce the waiting time cost of passengers, improve the scheduling efficiency, and the circuit of the vehicle under the condition of insufficient use of vehicle scheduling model for vehicle scheduling management, you can quickly restore the line spacing of the vehicle, and has certain practical value. Finally, according to the actual needs of the dynamic analysis of bus management benefits, put forward the analysis indexes of direct benefits and indirect benefits, discussed to reduce the error by loss, lower fares, travel time saving, bus sharing rate, the comprehensive benefits of transport costs cost savings and reduce traffic congestion caused by taking Xi'an city as an example, has carried on the case analysis, the calculation results show that the implementation of digital information management. The benefit is remarkable, and it shows a good prospect of application.

【学位授予单位】：长安大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：U491.17

【引证文献】