【作者】 张邻；

【导师】 杜文；

【作者基本信息】 西南交通大学 ， 交通工程， 2011， 博士

【摘要】 我国土地利用的半市场化、高密度开发、人口密集和快速城市化等现状特征与欧美等国的发展状况存在较大的差异,国外的土地利用模型大多数不完全适用于我国国情,且国外不同模型开发的目的和侧重点不同,对于我国而言,交通与土地利用一体化模型构思和实用化还有很长的路要走,需要研究适合高密度开发和快速城市化发展的机理模型,积累本土化研究经验,在理论上或规划模型方法上必须要获得突破。本论文立足我国城市发展实际,研究城市居民居住地选择与交通系统相互关系的理论与方法,为城市交通规划制定、交通政策分析以及城市空间结构调整提供理论基础。1为了检验交通与活动场所之间关系,通过建立房主与移动家庭之间的谈判模型,得到并证明了城市每个小区每期房价的NASH均衡解,每个移动家庭根据均衡房价、位置引力、广义交通费用得到各个小区的效用,通过模糊效用估计,获得候选小区集,并在候选集上采用非协作博弈的混合策略。在此基础上,提出了谈判机制混合策略选择居住地一定存在均衡解,并通过构造函数,利用Brouwer不动点定理进行证明。通过设计求解谈判机制模型的算法,并基于一个算例,分析了道路改善前后居民移动居住地的不同,给出了引导居民对居住地选择方法,针对交通影响居住地选择的特点,对城市结构完善提出了规划与管理建议。2建立住房估价模型,并通过建模探讨轨道交通与其它交通方式之间的相互衔接,获得家庭交通阻抗。每个移动家庭根据各小区的住房价格、享乐成本和家庭广义交通费用,获得各小区对自己的吸引力,并由贴近度得到各小区的满意度,移动家庭根据适当的满意度和价格,获得家庭剩余最大的居住地。3根据住房价格、家庭的广义交通阻抗和小区享乐设施的完备性,建立对各小区居住地进行满意度评估模型,并结合家庭收入和房价建立家庭消费剩余模型。通过对路网实行区域或警戒线拥挤征收政策,由家庭每个成员不同出行目的的旅行链特征和相应的出行方式,得到家庭的通勤广义交通阻抗和节假日娱乐出行的广义交通阻抗,并根据家庭消费剩余模型获得家庭剩余最大的居住地。最后,通过例证不同拥挤征收方式和征收水平,观察拥挤征收对居民居住地选择的影响。4为了在交叉口合理地分配通行权,通过代理人技术,获得各个相邻的代理人之间相互博弈模型,并建立衡量双方谈判效果的Nash均衡模型,证明了双方谈判的Nash均衡解的存在性,得到了各个相邻代理人之间相互博弈的均衡解,从而获得路网的均衡价格(费用)。根据路网的均衡价格,各个驾驶者采取费用总和最小的路径策略或模糊对比策略,选择对自己最有利的路径。结果表明：合同模式算法使得各交叉口的信号控制进行协调统一,充分发挥交通管理控制的主观能动性,并把管理者的主动性和驾驶者的个人理性行为结合起来,使得路网系统效用提高55.58%,是研究路网系统最优和用户最优之间相互协调有效方法。为了进一步提高整个路网的运行效率,采取对交叉口的信号控制方案进行优化策略。根据相邻交叉口流出交通流大小和方向不同,通过动态模糊控制手段,采用弹性信号周期和最大隶属度原则,合理地动态分配交叉口信号相位的绿灯时间。驾驶者通过对路径察觉,获得最小察觉阻抗路径,并与智能交通系统提供的各路径阻抗进行对比,选择最大满意度的路径。结果表明,动态模糊控制和决策法与最短路径法的交通流分配结果基本一致,模糊最优控制使得信号交叉口的控制效用进一步提高50%以上,且动态模糊控制和决策法更能反映人的行为,是一种有效的交通控制方法。更多还原

【Abstract】 China’s land use is half market, high-density development, population density and rapid urbanization the current characteristics which is a big different from Europe and the United States and other countries. Most of the foreign land-use models are not fully applicable to the conditions of our country, and the purpose and focus of the foreign development different models are different. For China, ideas and practical are still a long way to go about transport and land use the integration model. The mechanism model of high-density development and the rapid urbanization need to be studied. So localization experience is accumulated, and theory or planning model approach must be to a breakthrough. In this dissertation, some theories and methodologies about the relationship between dynamic traffic and residential location choice were proposed in accordance with the special characteristics of Chinese cities. This dissertation lays a solid theoretical foundation for urban transportation planning, traffic policy analysis, and urban spatial structure alteration.1 In order to test the relationship between transport and residential location, the negotiation model was proposed between the homeowner and mobile home, and NASH equilibrium solution of city each district housing prices was got and proven. The effectiveness of the various districts of each mobile home was got by a balanced price, the location gravity and the generalized transportation imprecise. The candidate cell set was received by the fuzzy effectiveness estimation, and a mixed strategy of non-cooperative game was adopted in the candidate cell set. The existence of the balanced solution was proposed to choose their residential location about hybrid strategy negotiation mechanism on the basis of this mixed strategy. A function was constructed, and the existence of the balanced solution was proven by Brouwer’s fixed point theorem. The algorithm of solving the negotiation model was proposed. Base on a numerical example, residential location choice was analyzed by the road improvement, and guidance method of the resident choice location is given. According to traffic impact residential choice features, structural perfection of urban planning and management suggestion is proposed. 2 A housing valuation models is established. The interface link is explored through the establishment of model between the rail transit and the other transport modes, and family transportation impedance is acquired. Each mobile home gets the attractiveness of the various districts for their own in accordance with a balanced price, hedonic cost and the generalized transportation impedance, and satisfaction of all residential is received by the close degree. In accordance with appropriate satisfaction and price, the mobile home obtains the greatest residence of consumer surplus.3 First, according to house prices, the family generalized transport impedance and the completeness of community hedonic facilities, the satisfaction assessment model is proposed to assess the residential location, and the household consumption remaining model is established by the satisfaction, household income and house prices. Second, cordon or area-road congestion charging regimes is implemented, and the generalized commuting traffic impedance and generalized holiday entertainment traffic impedance are got by different trip purpose travel chain characteristics and the corresponding travel mode of family members. The residence of the largest family remaining is got by the household consumption remaining model. Last, according to the different congestion charging methods and different charging levels, residential location choice is observed by a numerical example.4 In order to allocation the right-of-way of intersection reasonably, the game model was obtained between each adjacent agent by agent technology. The Nash equilibrium model of measurable negotiations effects was proposed, and the existence of Nash equilibrium solution was proven. The game equilibrium solution was got between each adjacent agent, so the equilibrium strategy price (cost) was obtained in the network. According to the equilibrium strategy price, travelers take the smallest prices path strategy or fuzzy comparison hybrid strategy in transportation networks, so their own best path was chosen. The results indicate that contract model algorithm makes the intersection signal control harmonize and shows subjective initiative of traffic control and management fully. The contract-base dynamic traffic game model combines with the management initiatives and the driver’s personal rational behavior, and the control effectiveness of road network system increases by 55.58% based on contract model algorithm, so it is an effective measurement for studying coordination between system optimal and user optimal. In order to improve the operational efficiency of the whole traffic network, intersection signal control program was optimized. According to magnitude and direction of the adjacent intersection traffic flow, intersection signal dynamic phase was reasonably allocated by the different elastic signal period and the maximum membership degree principle. Based on the observation path of the driver, the minimum detection impedance path was obtained, and the impedance was compared with the all path impedance of intelligent transportation system, so the greatest satisfaction path was chosen. The results indicate that traffic assignment of the dynamic fuzzy control and decision-making method is consistent with the shortest path method, and that signal intersection control effectiveness is futher increased 50% or more by the fuzzy optimal control, and dynamic fuzzy control and decision making could more accurately reflect human behavior and could be an effective measurement for that signal intersection control.更多还原