【摘要】： 隨著國民經(jīng)濟的發(fā)展，人們對各類交通的需求不斷增加。而交通流量的增長與現(xiàn)有道路狀況之間的矛盾日益突出，成為限制社會經(jīng)濟發(fā)展的主要因素之一。交通流理論的研究目標(biāo)是要建立能夠描述實際交通一般特性的交通流模型，揭示交通流的基本規(guī)律。開展交通流理論的研究，不僅有著深遠(yuǎn)的科學(xué)意義，而且具有重要的工程應(yīng)用價值。 論文在元胞自動機理論的基礎(chǔ)上，對交通流模型的建模、仿真及其應(yīng)用等方面進(jìn)行了研究，主要包括以下內(nèi)容： 1、對元胞自動機交通流模型進(jìn)行了分類、仿真以及分析評價。根據(jù)元胞自動機交通流模型中車輛占據(jù)的元胞個數(shù)、車道數(shù)量以及模型維數(shù)將其分為單細(xì)胞和多細(xì)胞TCA模型、單車道和多車道TCA模型、一維和二維TCA模型。在此分類的基礎(chǔ)上，，應(yīng)用Java程序?qū)γ糠NTCA模型的時空關(guān)系和交通流基本圖進(jìn)行了細(xì)致的仿真，并結(jié)合仿真結(jié)果對模型進(jìn)行了分析和評價。 2、建立了一種改進(jìn)的NS模型，并對模型的隨機特性進(jìn)行了研究和應(yīng)用。將一組恒速控制條件引入NS模型的速度控制規(guī)則之中，提出了一組新的速度控制規(guī)則。在此基礎(chǔ)上，將隨機減速概率p從新模型中提取出來，作為一個隨機參數(shù)植入一階連續(xù)介質(zhì)理論交通流模型(LWR模型)之中，并重構(gòu)了其交通流基本圖。數(shù)值模擬結(jié)果和實測數(shù)據(jù)對比表明，這種方法能夠較為真實的再現(xiàn)實際交通流特征，并且很好的詮釋了交通流的流體力學(xué)特性。 3、將車輛的OD屬性引入二維的BML模型中，根據(jù)系統(tǒng)內(nèi)車輛是否守恒建立了兩種二維TCA模型。定義了模型中車輛的起訖點、起訖距離以及起訖距離的測量方法，并給出了模型中起訖距離概率依從的三種分布方式，即指數(shù)分布、冪分布和均勻分布。在此基礎(chǔ)上給出了模型中車輛的基本運動規(guī)則，建立了車輛守恒的擴展BML模型和非守恒的擴展BML模型。對新建模型以及模型中參數(shù)之間關(guān)系進(jìn)行了數(shù)值模擬，結(jié)果表明：通過合理的選擇起訖距離的概率分布形式以及調(diào)節(jié)模型中參數(shù)的大小，可以提高城市道路的通行能力，減小車輛的出行時間。 4、基于元胞自動機的高效運算能力，結(jié)合出發(fā)時間選擇模型、動態(tài)路線選擇模型和動態(tài)網(wǎng)絡(luò)裝載模型，借鑒動態(tài)建模的思想，構(gòu)建了基于元胞自動機的動態(tài)交通分配系統(tǒng)框架。 5、在GIS技術(shù)支持下，根據(jù)以上構(gòu)建的動態(tài)交通分配系統(tǒng)框架，建立了一種用于動態(tài)交通分配的多模式、可擴展的路網(wǎng)數(shù)據(jù)結(jié)構(gòu)。將此數(shù)據(jù)結(jié)構(gòu)轉(zhuǎn)化為GIS中的屬性數(shù)據(jù)，以實際路網(wǎng)為背景進(jìn)行了動態(tài)交通分配，并根據(jù)分配結(jié)果對當(dāng)前路網(wǎng)做出了分析、評價和改進(jìn)建議。 最后，對論文的工作進(jìn)行了總結(jié)，并指出了下一步需要繼續(xù)研究的內(nèi)容。
[Abstract]:With the development of national economy, people's demand for all kinds of traffic is increasing. The contradiction between the growth of traffic flow and the existing road condition is becoming more and more prominent, which has become one of the main factors restricting the development of social economy. The research goal of traffic flow theory is to establish a traffic flow model which can describe the general characteristics of real traffic and to reveal the basic laws of traffic flow. The research of traffic flow theory not only has profound scientific significance, but also has important engineering application value. Based on the theory of cellular automata, this paper studies the modeling, simulation and application of traffic flow model. The main contents are as follows: 1. The traffic flow model of cellular automata is classified. Simulation and analysis and evaluation. According to the number of cells occupied by vehicles, the number of lanes and the dimension of the model, the model can be divided into single cell and multi-cell TCA model, single-lane and multi-lane TCA model, one-dimensional and two-dimensional TCA model. On the basis of this classification, the spatio-temporal relationship of each TCA model and the basic traffic flow diagram are simulated with Java program, and the model is analyzed and evaluated with the simulation results. 2. An improved NS model is established, and the stochastic characteristics of the model are studied and applied. A set of constant speed control conditions is introduced into the speed control rules of NS model, and a new set of speed control rules is proposed. On this basis, the random deceleration probability p is extracted from the new model as a random parameter and implanted into the first-order continuum theoretical traffic flow model (LWR model), and its basic traffic flow graph is reconstructed. The comparison between the numerical simulation results and the measured data shows that this method can reproduce the real traffic flow characteristics and explain the hydrodynamic characteristics of the traffic flow very well. 3. The OD attribute of the vehicle is introduced into the two-dimensional BML model, and two two-dimensional TCA models are established according to whether the vehicle in the system is conserved or not. In this paper, the methods of measuring the starting point, starting distance and starting distance of vehicle in the model are defined, and three distribution modes of probability compliance of starting and end distance in the model are given, that is, exponential distribution, power distribution and uniform distribution. On this basis, the basic motion rules of the vehicle in the model are given, and the extended BML model of vehicle conservation and the extended BML model of non-conserved vehicle are established. The new model and the relationship between the parameters in the model are numerically simulated. The results show that the traffic capacity of urban roads can be improved by selecting the probability distribution form of the starting distance and adjusting the size of the parameters in the model. Reduce vehicle travel time. 4. Based on the efficient computing ability of cellular automata, combined with the departure time selection model, the dynamic route selection model and the dynamic network loading model, the idea of dynamic modeling is used for reference. The framework of dynamic traffic assignment system based on cellular automata is constructed. 5. With the support of GIS technology, according to the above framework of dynamic traffic assignment system, a multi-mode and extensible road network data structure for dynamic traffic assignment is established. The data structure is transformed into attribute data in GIS, and dynamic traffic assignment is carried out under the background of the actual road network. According to the result of the distribution, the analysis, evaluation and improvement suggestions are made for the current road network. Finally, the paper summarizes the work, and points out the next step to continue to study the content.
【學(xué)位授予單位】：山東科技大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2007
【分類號】：U491

【引證文獻(xiàn)】

相關(guān)期刊論文 前1條

1 陳永;王曉明;黨建武;;基于元胞自動機理論的快速公交系統(tǒng)能耗研究[J];計算機應(yīng)用;2012年03期

相關(guān)博士學(xué)位論文 前3條

1 陳晨;基于細(xì)胞自動機方法的城市交通流模擬及優(yōu)化[D];華中科技大學(xué);2010年

2 柯姜岑;基于元胞自動機的水運樞紐運輸組織研究[D];武漢理工大學(xué);2012年

3 袁紹欣;城市交通擁堵傳播機理及其控制策略研究[D];長安大學(xué);2012年

相關(guān)碩士學(xué)位論文 前2條

1 吳國輝;北京站交通銜接仿真與優(yōu)化研究[D];北京交通大學(xué);2011年

2 宋升;基于元胞自動機的集裝箱碼頭前沿路網(wǎng)結(jié)構(gòu)建模與分析[D];武漢理工大學(xué);2011年

本文編號：2303716