交通系统的非线性分析、控制及系统实现

发布时间：2018-04-10 17:30

  本文选题：交通流 + 分支分析　； 参考：《西北工业大学》2015年硕士论文

【摘要】：交通控制是缓解交通拥堵的重要手段之一。论文通过结合实际交通应用背景研究道路拥堵问题,在分析交通拥堵产生机理的基础上设计拥堵控制策略来代替人工调节控制,并从工程化角度设计实现小型化的交通流信息在线处理系统来实时监控道路交通的运行状况,从而试图获得一种缓解交通拥堵的有效方法。论文主要工作总结如下:(1)针对现有交通微观模型对实际道路车辆运行状况描述不足的问题,通过分析驾驶员心理和车辆动力学特征对优化速度模型进行改进,结合实测数据对改进后的模型进行验证并进行模型稳定性分析,结果表明改进后的模型能够更好地对实测数据进行拟合,可对实际交通运行状况进行进一步地准确描述。(2)针对现有非线性分析方法对交通流应用的局限性,将分支理论引入交通模型的非线性分析中,通过分析模型分支条件以及分支点邻域内的交通状态变化来研究交通拥堵的作用机理。其中,微观模型的分支分析采用传统针对常微分方程的Hopf分析方法,而针对宏观模型的偏微分方程表达形式,则是通过引入两个等价的变量代换,如同将离散系统单位圆内的稳定性扩大到整个复平面,将交通流问题转化为系统稳定性问题,然后再针对模型平衡点分支进行重点分析。结果表明,分支的出现是导致交通失稳进而演化为交通阻塞现象的一个重要原因。(3)针对交通失稳现象,分别以微观改进优化速度模型和宏观速度梯度模型为例来探讨缓解交通拥堵的控制方法。其中,针对微观模型的反馈控制方案,在理论分析反馈系数对控制模型稳定性影响的基础上进行MATLAB仿真,结果表明通过合理调整反馈系数可有效缓解交通拥堵现象;针对宏观模型的入口匝道模糊控制方案,构造了以当前路段偏离期望密度的差值和相邻路段的密度差为输入量、以匝道调节率为输出量的二维模糊控制器,MATLAB仿真分析的结果表明,该控制方案对拥堵路段的交通状况具有明显改善作用。(4)从工程应用角度搭建基于FPGA的交通流信息在线处理系统,完成具有4片FPGA阵列结构的6层电路板的PCB设计及硬件电路调试,基于QuartusⅡ开发工具完成交通流模型的FPGA仿真验证,并以宏观模型为例进行了系统整体功能的测试,结果表明系统能够在线处理交通流数据并将数据结果通过显示界面实时、直观地显示出来,满足设计功能要求。
[Abstract]:Traffic control is one of the important means to alleviate traffic congestion.Based on the analysis of the mechanism of traffic congestion, a congestion control strategy is designed to replace manual control.A miniaturized online traffic flow information processing system is designed and implemented from the engineering point of view to monitor the traffic condition in real time so as to obtain an effective method to alleviate traffic congestion.The main work of this paper is summarized as follows: (1) aiming at the problem that the existing traffic microscopic models do not describe the actual road vehicles, the optimization speed model is improved by analyzing the drivers' psychology and vehicle dynamics characteristics.The improved model is verified with the measured data and the model stability is analyzed. The results show that the improved model can fit the measured data better.In view of the limitation of the existing nonlinear analysis method to traffic flow, branch theory is introduced into the nonlinear analysis of traffic model.The mechanism of traffic congestion is studied by analyzing the model bifurcation condition and the traffic state change in the neighborhood of the branch point.Among them, the branch analysis of microscopic model adopts the traditional Hopf analysis method for ordinary differential equation, and the partial differential equation for macroscopic model is expressed by introducing two equivalent variables.As the stability in the unit circle of discrete system is extended to the whole complex plane, the traffic flow problem is transformed into the stability problem of the system, and then the bifurcation of the equilibrium point of the model is analyzed emphatically.The results show that the emergence of branches is an important reason that leads to traffic instability and then evolves into traffic congestion.The control methods to alleviate traffic congestion are discussed by taking the microscopic improved optimal speed model and the macroscopic velocity gradient model as examples respectively.According to the feedback control scheme of micro model, MATLAB simulation is carried out on the basis of theoretical analysis of the influence of feedback coefficient on the stability of the control model. The results show that the phenomenon of traffic congestion can be effectively alleviated by adjusting the feedback coefficient reasonably.Aiming at the fuzzy control scheme of the entrance ramp in the macro model, the difference between the current road section deviation from the expected density and the density difference between adjacent sections is constructed as the input.MATLAB simulation results of 2-D fuzzy controller with ramp regulation rate as output show that,This control scheme can obviously improve the traffic condition of congested sections. (4) from the point of view of engineering application, an on-line traffic flow information processing system based on FPGA is built.The PCB design and hardware circuit debugging of the six-layer circuit board with four FPGA arrays are completed. The FPGA simulation verification of the traffic flow model is completed based on the Quartus 鈪，

本文编号：1732213