大跨桥梁多车道随机车流统计及其等效荷载模型研究

发布时间：2018-08-08 15:27

【摘要】：随着桥梁跨度的不断增大、车流量的不断增加、车速的不断提升以及重车及超重车的不断出现等众多因素,作用在桥梁上的随机车流变得越来越复杂。由于大跨桥梁上的随机车流荷载具有较强的随机性,传统的假定车间距及车速不变的研究方法具有较大的局限性,难以对随机车流荷载的特性进行准确的分析。本文基于南京长江二桥收费站取得的随机车流数据结合实测的车流数据,统计了二桥各类车型的车重特征参数,研究了大跨桥梁各个不同车道的在不同交通服务水平下的车型比例以及车辆到达率等特征参数;基于元胞自动机模型(CA模型)的基本思想,从交通控制理论的单车道元胞自动机模型入手,并结合实测的各车道车型比例及断面发车率等断面发车数据以建立多车道随机车流模型的数学模型,依托Matlab开发了多车道随机车流模拟程序,以对二桥的随机车流进行仿真模拟,并对随机车流运行过程中的车流特性进行了研究。主要包括以下几个方面的内容：(1)研究车辆在桥梁上的行驶规律,包括加减速原则以及车辆在行驶过程中换道规律,确定车辆在需要换道时应如何换道以及换道的概率。根据车流的运行规则并基于CA模型思想建立大跨桥梁多车道随机车流的数学模型。(2)基于建立的随机车流的数学模型并依托Matlab软件开发大跨桥梁多车道随机车流模拟程序,以对二桥的随机车流进行仿真模拟,利用Matlab的界面设计功能以及句柄图形功能进行模拟程序的输出界面设计,并实现随机车流模拟的可视化。该程序的实现为得到随机车流荷载数据提供了支持。(3)从南京长江二桥收费站以及通过对随机车流进行直接观测统计获取大量随机车流交通特征参数,包括车型比例参数、各个车道在不同交通服务水平下的断面发车率以及不同车型车重数据分布参数；对本文交通调查所取得的断面发车率参数样本的分布类型进行了卡方拟合优度检验,发现在不同交通服务水平下三个车道的断面发车率基本服从泊松分布。(4)基于二桥的实测数据以及元胞自动机模型理论对大跨桥梁的随机车流进行模拟,将模拟车流特征参数与实测车流数据进行了对比分析,从而验证程序的准确性及合理性；研究不同交通服务水平下车流的运行特点；此外,进一步对模拟随机车流下的交通量、车速以及车间距进行规律研究及特征分析,发现三种交通服务水平下的车间距都基本服从对数正态分布,拟合出了相应的对数正态分布曲线,并通过K-S检验对其进行了验证。(5)通过对随机车流的模拟可以得到每个时刻桥面上车辆的位置、对应的车型、车重以及车速等参数。引入等效动态车辆荷载方法(EDWL法),利用该方法将随机车流位置转换为了随机车流荷载,简化车桥之间的相互作用,并生成桥梁部分断面的荷载时程曲线。
[Abstract]:With the increasing span of bridges, the increasing of traffic flow, the increasing speed of vehicles, and the emergence of heavy and overweight vehicles, the random traffic flow on the bridge becomes more and more complex. Because of the strong randomness of the random vehicle flow loads on long-span bridges, the traditional research methods which assume that the vehicle spacing and the speed are invariant have great limitations, so it is difficult to accurately analyze the characteristics of the random vehicle flow loads. Based on the random traffic flow data obtained from Nanjing No. 2 Changjiang River Bridge toll station and the measured traffic flow data, the characteristic parameters of vehicle weight of all kinds of vehicle types in the second bridge are calculated in this paper. Based on the cellular automata model (CA model), the paper studies the vehicle model proportion and vehicle arrival rate of different lanes of long span bridge under different traffic service levels. Starting with the single lane cellular automata model of traffic control theory and combining with the measured data of each lane model and section departure rate, the mathematical model of multi-lane stochastic vehicle flow model is established. Based on Matlab, a multi-lane stochastic vehicle flow simulation program is developed to simulate the random traffic flow of the second bridge, and the characteristics of the vehicle flow in the course of the random traffic flow are studied. The main contents are as follows: (1) study the driving law of vehicle on the bridge, including acceleration and deceleration principle and the law of vehicle changing in the course of driving, and determine how to change the road and the probability of changing the road when the vehicle needs to change the road. According to the running rules of vehicle flow and the idea of CA model, the mathematical model of multi-lane random vehicle flow of long-span bridge is established. (2) based on the established mathematical model of random traffic flow and relying on Matlab software, the simulation program of multi-lane random vehicle flow of long-span bridge is developed. The random traffic flow of the second bridge is simulated, the output interface of the simulation program is designed by using the interface design function of Matlab and the function of handle figure, and the visualization of the simulation of random traffic flow is realized. The implementation of the program provides support for obtaining random vehicle flow load data. (3) A large number of traffic characteristic parameters of random traffic flow, including model scale parameters, are obtained from Nanjing Changjiang River Bridge toll station and through direct observation and statistics of random traffic flow. Section departure rate and vehicle weight data distribution parameters of each lane under different traffic service levels. The distribution types of cross-section departure rate parameters obtained in this paper are tested by chi-square goodness of fit. It is found that the section departure rate of three lanes is distributed from Poisson under different traffic service levels. (4) based on the measured data of the second bridge and the cellular automata model theory, the random traffic flow of the long-span bridge is simulated. The characteristic parameters of simulated traffic flow are compared with the measured traffic flow data to verify the accuracy and reasonableness of the program; to study the running characteristics of the traffic flow under different traffic service levels; in addition, to further study the traffic volume under the simulation of random traffic flow, The law of vehicle speed and vehicle spacing is studied and the characteristics are analyzed. It is found that the distance between vehicles under the three traffic service levels is basically applied to the logarithmic normal distribution, and the corresponding logarithmic normal distribution curve is fitted out. It is verified by K-S test. (5) through the simulation of the random traffic flow, the position of the vehicle on the bridge deck at each moment, the corresponding vehicle type, the vehicle weight and the speed of the vehicle can be obtained. The equivalent dynamic vehicle load method (EDWL method) is introduced, by which the random vehicle flow position is transformed into the random vehicle flow load, which simplifies the interaction between the vehicle and the bridge, and generates the load time history curve of some sections of the bridge.
【学位授予单位】：南京工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491;U441.2

【相似文献】