【作者】 慈玉生；

【导师】 裴玉龙；

【作者基本信息】 哈尔滨工业大学 ， 道路与铁道工程， 2008， 博士

【摘要】 城市快速路是我国大、中城市道路网系统的重要组成部分,是缓解城市交通拥挤的重要举措之一。然而,城市快速路建设投资巨大、涵盖范围较广,引导着城市的发展方向,所以快速路的建设必须科学、合理。随着我国城市快速路建设步伐的加快,对快速路的规划、建设及运营管理的要求不断提高,现有体系已无法满足需求,加快快速路规划、建设及运营管理相关研究显得尤为迫切。论文以教育部高等学校博士学科点科研专项基金为依托,以快速路入口、出口匝道连接段为研究对象,在进行大规模交通调查的基础上,运用统计分析理论、元胞自动机理论、交通波理论、模糊控制理论等相关理论与方法,开展快速路匝道连接段交通流运行特性、通行能力计算方法及匝道设置相关研究,建立并确定快速路匝道连接段通行能力及匝道设置指标体系。通过这些方法和模型的研究,以期为我国快速路系统的规划、建设及运营管理提供科学依据,同时也为其它高速道路提供重要借鉴。论文选用磁映像及气压管脉冲等现代交通流检测技术,采用组合检测方法进行调查方案的设计和实施,获取了我国典型大、中城市快速路入口、出口匝道连接段交通流、道路基础设施及交通管理措施等方面的数据。通过对匝道连接段总体交通运行特性分析的基础上,结合调查数据着重分析了匝道连接段交通流流量速度特性、车头时距特性、加减速度特性及宏观交通流三参数特性,从而建立了匝道连接段车头时距分布的威布尔模型及宏观交通流三参数的相互关系模型。为后续匝道连接段通行能力计算模型及匝道设置研究奠定了基础。针对当前匝道连接段通行能力的确定基本沿用美国《HCM》方法存在的不足,论文采用间隙接受理论、交通波理论及元胞自动机理论等分别建立了快速路入口、出口匝道连接段的通行能力的计算模型。在对入口、出口匝道连接段的实测数据分析和处理的基础上,对建立模型的相关参数进行标定,从而确立了城市快速路入口、出口道连接段的通行能力值,并给出了相应的通行能力推荐值。在入口、出口匝道连接段通行能力计算方法研究的基础上,分析了快速路入口匝道控制的类型及控制效果,总结并归纳了三种入口匝道控制策略,即主线优先控制策略、交替通行策略和信号控制策略,分析了各种控制策略的优缺点,研究确定了入口匝道调节条件下的匝道连接段通行能力。针对现有快速路匝道变速车道设置条件不确定性的不足,提出了快速路变速车道的设置条件。在此基础上,利用车辆动力学理论、间隙接受理论建立了快速路变速车道长度计算模型,进一步研究集散车道和匝道间距设置方法问题,给出了城市快速路主线不同设计速度下的变速车道长度及相邻匝道间距的建议值,并给出了集散车道设置原则及集散车道长度计算方法。结合模糊理论和神经网络理论的优点,提出了基于神经模糊系统控制法来实现快速路入口匝道的自适应控制,并给出了控制器的设计步骤和实例,结果表明入口匝道控制能有效提高快速路入口匝道连接段的通行能力。最后,利用VISSIM微观仿真软件构建了入口、出口匝道连接段的仿真模型,并结合实测数据对模型参数进行了标定,给出了仿真结果。同时,利用Matlab相关工具箱开发了快速路入口、出口匝道连接段的元胞自动机仿真模块,并进行了不同变速车道长度的快速路入口、出口匝道连接段交通流仿真,从而有效验证了匝道连接段通行能力计算模型及相关设置的理论研究成果。更多还原

【Abstract】 Urban expressway is an importantcomponentof the roadnetwork system inlarge- and medium-sizedcities in China, and it’s to be approved it’s one of keymeasurements to release traffic congestion. However, urban expressway is a hugestructure with extensive investment, so the construction mustbe scientific andrational, which will guide to the city’s developing direction. With rapid construction,the differentdemands are continually increasing. It’s particularly urgentto do somerelevantresearchs onplanning, construction, operation and managementforurbanexpressway, because of the existing systemlagging farbehind thandemand.On- andoff-ramp junctions of urban expressway were chosen as two mainresearch objects depending on Research Fund forthe Doctoral Programof HigherEducation. On the basis of a large-scale traffic survey, statistical analysis theory,cellularautomaton theory, traffic wave theory and fuzzy control theory were used toanalyze and study the traffic flow operationcharacteristics, capacity computingmethods and ramp settings. Fromabove study, we hoped to find some scientic basison planning, construction, operation and managementforChinese urban expressway,as well as forotherhigh-speed roads orhighways.Combined detection methods were used to design traffic flow surveyprogramme with the modern magnetic imaging and metrocount impulse technologies.Aftersurvey application, representative urban expressway on- and off-ramp data fortraffic flow, roadway infrastructure, traffic management and control, and etc werecollected in China. Based on analyzing general ramp junctions’traffic flow operationcharacteristics, different characteristics for traffic volume and speed, headway,acceleration and deceleration, and density were studied according to collected data.Consequently, Weibull distribution model for headway and relationship models formacroscopic traffic flow parameters were set up. Those were a good foundation forestablishing on- andoff-rampjunctions computing models andrampsetting research.Highway Capacity Mannual of America was widely used to compute andanalyze capacity in China. Actually, there were some disadvantages someway. On-and off-ramp junctions computing methods were separately set up according to time gap acceptace theory, traffic wave theory, and automaton cellular theory. And then,the relevant parameters calibration forthose models was done according to collecteddata. So the possible capacity values were determined forChinese urban expresswayon- and off-ramp junctions. Finally, suggested possible capacity vaules for on- andoff-ramp junctions of Chinese urban expressway were put forward aftercomparisonanalysis on differentvalues.Further more, on-ramp control types and application effects were attentivelystudied depend on the computing capacity methods. Subsequently, three on-rampmetering strategies were concluded, such as mainroad priority strategy; alternatepassing through strategy, and signal control strategy. On this basis, on-ramp junctioncapacities were studied in condition of different control strategy after comparasinganalysis the advantages anddisadvantages.The new setting condtions forurban expressway ramp were put forward aimingat the uncertainty of present speed changing lanes setting conditions. On this basis,the length computing models for speed changing lanes were set up,collector-distributor lane and ramp distance setting methods were studied and therelevant suggested values ormethods were put forward separately. Combination themerits between fuzzy theory and neural network theory, a self-adaptive on-rampmetering method for urban expressway was put forward according to neuro-fuzzytheory, and also for the design programe for the controller and an example. Theexperiment results indicated that on-ramp junction capacity increased much more incondition of metering.Finally, the simulation models for on- and off-ramp junctions were built up inVISSIM software environment, the parameters calibration were done with collecteddata, and the simulation results were listed. At the same time, automaton cellulartraffic flow simulation modulars foron- and off-ramp junctions were developed withMATLAB toolboxes, and the simulation experiments were done with the differentlengths of speed changing lanes. The computing models and relevant ramp settingresearch findings were tested by the simulation results.更多还原