随机环境下公交系统鲁棒性优化方法研究

发布时间：2018-01-10 09:18

本文关键词：随机环境下公交系统鲁棒性优化方法研究　出处：《吉林大学》2017年博士论文　论文类型：学位论文

【摘要】：城市公交系统作为与城市交通系统和城市居民出行相关联的开放式复杂系统,多种随机因素都可能会造成公交系统的无序运营。随着社会经济的快速发展,城市公交作为城市居民主要的出行方式之一,公交运营的可靠性和稳定性受到更多关注,本文将其统称为公交系统的“鲁棒性”,主要体现为城市单线路公交运营的鲁棒性。城市公交系统中包含大量的随机扰动因素,如行程速度的不确定性、公交站点乘客到达的随机性等,公交车在运行过程中由于受随机交通扰动影响无法准确遵循既定的调度计划运行,致使公交系统运行稳定性和服务质量下降。本文针对城市公交系统随机扰动诱发的公交无序运营问题,以常规公交系统频发的公交串车为研究对象,对公交串车情况下公交车辆对道路交通运行影响分析,构建基于交通波理论的车辆附加延误计算模型,并针对公交客流到达公交站点的随机波动性,以二维时间序列上公交站点的客流数据统计与分析为基础,设计基于二维时间序列的公交站点客流预测模型,进而分别以公交驻站控制和协调控制为手段,以自均衡车头时距为控制策略,设计随机环境下公交系统鲁棒性优化方法,提升公交系统抵御随机干扰的能力。本文通过对公交系统控制中各重要组成部分的探讨和研究,形成了系统、科学的城市常规公交运营实时控制体系。本文主要工作内容和研究成果概括如下。(1)基于交通波理论的车辆附加延误计算方法研究。针对直线式公交站点布设在道路交叉口附近时公交串车对道路车辆运行影响问题,首先,以单一公交车在公交站点停靠对道路交通流运行影响特性剖析为基础,依据公交车驶入研究路段的时刻不同,将公交车对道路车辆延误影响划分为六种情况,以交通波理论为理论工具,分类别构建单一公交影响下道路车辆附加延误计算模型;进而,剖析直线式公交站点设置单泊位和双泊位时,聚集到达站点的两辆公交车运行机理,并分别针对公交站点单泊位和双泊位情况,设计公交车在站点处的停靠时间计算方法,以获取公交串车情况下,两辆公交车在站点处的停靠时间与道路车辆附加延误之间的关系,进而构建不同泊位下道路车辆附加延误计算模型。本部分的研究成果为公交实时控制方法研究提供必要的理论支撑。(2)基于二维时间序列的短时公交客流预测方法研究。公交客流到达公交站点的波动性是本文考虑影响公交系统运营稳定性的主要随机因素。综合考虑城市道路公交系统优化方法对短时公交客流的预知性需求,以二维时间序列公交客流数据(实时客流数据和历史客流数据)为数据基础,以卡尔曼滤波算法为模型基础,采用并行预测架构和决策级融合手段,设计基于二维时间序列的公交客流预测算法流程;以此为基础,以动态预测自适应权重系数为核心思想,以滚动式预测为预测模式,在考虑模型简洁性、预测可靠性等需求的基础上,分别设计了三种嵌入式公交客流预测决策级融合方法,为具有不同公交客流预测需求的控制系统提供短时客流预测方法支撑。本部分的研究成果能够为随机环境下城市公交系统鲁棒性优化方法体系提供准确的数据支持。(3)基于驻站控制的公交系统鲁棒性优化方法研究。在描述公交系统中公交车运营过程以及明确影响公交正常运行主要随机因素的基础上,探析公交系统中公交串车产生机理,以此为基础,对公交系统进行数学抽象,构建公交系统描述体系;针对公交系统受随机扰动干扰后恢复至稳定状态能力较差的问题,以提高公交系统车头时距收敛速度和减少控制点处驻站时间为目标,以自均衡车头时距为设计理念,提出一种相邻公交车车头时距自均衡控制策略;公交系统随机扰动因素较多,直接构建解析模型形式的公交驻站控制模型较为困难,因此,在对公交站点客流对公交运行影响分析的基础上,以逐步添加扰动因素为研究脉络,对控制点处的公交车进行实时控制,利用由简至繁逐级构建控制模型的方法,建立公交车驻站时间计算模型,在此基础上设计基于驻站控制的公交系统鲁棒性优化方法,并分别从理想情况、线性和非线性假设以及实际情况三个层面验证控制方法的有效性。本部分的研究成果为提升公交系统抵抗随机干扰能力提供新的解决思路。(4)基于协调控制的公交系统鲁棒性优化方法研究。针对不同客流需求下公交系统鲁棒性优化方法的适应性问题,在对公交跳站和公交驻站控制手段优缺点分析基础上,对公交驻站控制与公交跳站运营手段互补特性分析,并以公交跳站与公交驻站为协调控制手段,以自均衡公交车车头时距为协调控制策略,根据控制点处相邻公交车车头时距动态选择公交驻站或公交跳站,构建协调控制算法流程;通过对各控制手段下公交车车头时距迭代演化趋势预测,分别建立公交驻站和公交跳站动态选择模型,以此为基础,构建公交驻站时间计算模型,提出基于公交驻站和公交跳站协调控制的公交系统鲁棒性优化方法,并利用数值分析手段对协调控制中各种控制手段的选择情况进行分析,通过提取各控制手段的选择次数以及对应公交系统状态,分析协调控制方法对公交系统控制影响的内在机理,进而对协调控制方法适用范围进行讨论。本部分的研究成果为城市公交系统智能控制体系构建提供科学的控制方法支撑。本论文研究过程中以方法适用性、模型简洁性、理念先进性为主导思想,针对城市常规公交系统存在的公交串车问题,提出随机环境下公交系统鲁棒性优化方法,将先进的科学方法同城市公交系统运营实际相结合。本论文涉及的研究思想、脉络、方法以及研究成果对完善城市公交系统运营管理体系建设和丰富城市公交系统规划理论与方法体系具有重要的理论价值和实际意义。
[Abstract]:As the city bus system and city traffic system and city residents travel associated with the open complex system, disorderly operation many random factors are likely to result in the public transportation system. With the rapid development of social economy, city bus as one of the main city residents travel mode, bus operation reliability and stability is more attention, the the bus system referred to as the "robustness", mainly reflected the robustness of the single line operation of the city bus. A large number of random disturbance factors including city public transportation system, such as the speed of the uncertainty of passengers in the bus station at random, bus during operation due to random traffic disturbance can not accurately follow the schedule the operation plan established, resulting in the decrease of bus running stability and service quality. In this paper the city bus system random disturbance The bus operation disorder induced by bus, the bus transit system frequent train as the research object, analysis of the impact on the bus transit train case of road traffic, construction calculation model of vehicle based on the additional delay of traffic wave theory, and stochastic volatility of the bus station in the bus passenger flow, passenger flow and analysis to the data of two-dimensional time series of bus stations on the statistical basis of design of bus station passenger two-dimensional prediction model based on time series and then to bus station control and coordinated control as a means to self balanced car head pitch control strategy, design under random environment bus system robust optimization method to enhance the ability to resist the bus system random interference. Based on the bus in the control system of the important part of the discussion and the research, formed a system, city bus operation of Science When the control system. The main contents and research results are summarized as follows. (1) research on the calculation method of additional delay of vehicle traffic wave theory based on linear bus. According to the site layout at the intersection near the bus on vehicle impact on road vehicle operation problems, first of all, with a single bus in the bus stop sites of road traffic flow the influence of operation characteristics analysis based on different time sections of the research into the bus, bus delay on road vehicles will influence divided into six types, with the traffic wave theory, classified construction of single bus road vehicles under the influence of additional delay calculation model; then, analysis of linear single berth and double bus stations berth, gathered to the site two bus operation mechanism, according to single and double berth berth bus station, bus in site design The calculation method of the stopping time, to get on bus car, two bus at the site of the docking time between road vehicles additional delay, and the construction of road vehicles under different berth additional delay calculation model. The results of this part provide the necessary theoretical support for the real-time control method of bus (2). Study on the prediction method of short term passenger flow based on two dimensional time series. Bus passenger arrival fluctuation of the bus station is mainly considering influence of random factors transit system operation stability. Considering the city road transportation system optimization method to predict short-term bus passenger demand to the passenger data of two-dimensional time series (real-time data and historical traffic flow data) as the basic data, using Calman filtering algorithm using parallel architecture model based prediction and decision level fusion method. Prediction of passenger volume flow meter algorithm based on two-dimensional time series; based on dynamic prediction adaptive weight coefficient as the core idea, to predict rolling prediction model, considering the model simplicity, reliability prediction based on the requirement of design decision level fusion method for prediction of three kinds of embedded bus passenger is different bus passenger flow prediction control system needs to provide a short passenger flow forecasting support. Research results of this part can provide data support for the random environment city bus system robust optimization method system. (3) research of optimization method of bus station based on robust control system. On the base of describing the process of bus transit system and the normal operation of the main random factors influencing the bus on the bus on the car of the generation mechanism of public transportation system, which based on The bus system of mathematical abstraction, to build public transportation system description system; in the bus system are subjected to random disturbances after the return to steady state ability is poor, in order to improve the transit headway convergence speed and reduce the control point in time as the goal, to balance the headway as the design concept, put forward a kind of adjacent bus the headway of self balancing control strategy; bus system random disturbance factor more direct construction of bus station in the form of analytical model of control model is difficult, therefore, based on the analysis of the ring bus station passenger bus operation shadow, with disturbance factors as the research context gradually added, the real-time control of the control point of the bus the construction method of control model, from simple to complex step by step, the bus station set up time calculation model, based on the design of the bus in the station control system based on Method of robust optimization, and separately from the ideal case, the effectiveness of linear and nonlinear assumptions and the actual situation of the three level to verify the control method. The research results of this part provide a new solution for improving the bus system resistance to random interference. (4) the research of optimization method of bus system Robust Coordinated Control Based on adaptive. The problem of different passenger demand under the bus system robust optimization method, in the bus station and bus station jump control means on the basis of the analysis of the advantages and disadvantages of the bus station, bus station control and jump operation means complementary characteristic analysis, and to the bus station and the bus station jump to coordinate control means, to balance the bus the headway for coordinated control strategy, according to the adjacent control points of the bus headway in the dynamic choice of bus station or bus station to jump, construction of coordinated control algorithm based on the control process; The bus system means the headway iterative evolution trend prediction model was established to choose dynamic transit stations and bus station in the jump, on the basis of constructing bus station time calculation model, and puts forward the bus system robustness optimization method of bus station and bus station based on hop control coordination, and coordination of various control options the control means are analyzed by means of numerical analysis, by extracting each control means select the number and the corresponding bus system, analysis of the coordinated control method to control the inherent mechanism of the effect of the bus system, and the coordination control method applicable scope are discussed. The research results of this part provide scientific support for the control system of intelligent city bus the control system with the applicability of this model in the research process, simplicity, advanced concept as the guiding ideology, the city often The bus transit system on vehicle problems, put forward a random environment transit system robust optimization method, the advanced scientific methods to city transit system operation practice. The research thought, this thesis involves context, method and the research results of this dissertation have important theoretical value and practical significance to improve the city public transportation system construction and operation management system and abundant city public transportation system planning theory and method system.

【学位授予单位】：吉林大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：U491.17

【相似文献】