城市快速路交通状态估计与控制研究

发布时间：2018-07-28 13:50

【摘要】：城市快速路是全部或部分采用立体交叉与控制出入，供车辆以较高的速度行驶的道路，快速路的修建在一定程度上为缓解城市交通拥堵做出了贡献。随着交通出行的日益增加，快速路也出现了交通拥挤现象，而且愈加严重。为解决快速路交通拥挤问题，需要对快速路的交通状态进行估计，以全面掌握快速路的交通情况，为快速路交通流的管理、控制、诱导等提供有效的数据支撑。快速路交通状态是借助于交通流的相关参数来衡量某种交通状态的近似范围的，所以可以采用对交通状态参数进行估计的方式，间接实现对交通状态的估计。从本质上讲，交通状态参数估计是一个非线性、非高斯问题，鉴于粒子滤波在处理该类问题时，具有较好的优越性，本研究将快速路二阶宏观交通流模型与粒子滤波算法有效结合，建立交通状态参数估计模型，借助于MATLAB软件平台，实现对交通状态参数的准确估计。结果表明，基于基本粒子滤波构建的交通状态参数估计模型可以对快速路交通状态参数做出较好的估计。在粒子滤波算法应用过程中，会出现粒子退化现象。粒子退化现象的发生，不仅使得粒子多样性丧失，而且将大部分的计算集中到了一些对结果贡献很小的粒子上。为避免粒子退化现象的发生，本研究采用免疫粒子群算法对粒子滤波进行优化，运用改进后的粒子滤波算法进行实例验证，并与基本粒子滤波算法比较，结果表明，，改进后的粒子滤波算法可以实现对交通状态参数更精确的估计。交通状态本身具有一定的模糊性和不确定性，适合用模糊理论对其进行划分，本研究采用模糊C均值聚类方法将交通状态划分为畅通、轻度拥挤和拥挤三种。根据交通状态参数估计结果，将其聚类到相应的状态中，得到交通状态估计结果。在对快速路交通状态准确估计的基础上，可以实现对快速路主线及其匝道的有效控制，以最大发挥快速路的服务能力。将可变限速控制和入口匝道控制相结合，考虑不同交通状态下的特性，建立快速路联合控制模型，采用遗传算法对模型进行求解，并对其进行实例验证及应用。结果表明，联合控制模型可以较好的实现对快速路的有效控制。
[Abstract]:Urban expressway is a kind of road which is used in all or part of it to cross and control the entrance and exit of vehicles at a high speed. To some extent, the construction of expressway has contributed to the alleviation of urban traffic congestion. With the increasing of traffic travel, traffic congestion also appears on the expressway, and it is becoming more and more serious. In order to solve the problem of expressway traffic congestion, it is necessary to estimate the state of expressway traffic, so as to master the traffic situation of expressway and provide effective data support for the management, control and induction of expressway traffic flow. Expressway traffic state is based on the relative parameters of traffic flow to measure the approximate range of traffic state, so the estimation of traffic state parameters can be used to indirectly realize the estimation of traffic state. In essence, traffic state parameter estimation is a nonlinear, non-Gao Si problem. In view of the superiority of particle filter in dealing with this kind of problem, In this paper, the second order macroscopic traffic flow model of expressway and particle filter algorithm are combined effectively, and the traffic state parameter estimation model is established, and the accurate estimation of traffic state parameter is realized by means of MATLAB software platform. The results show that the traffic state parameter estimation model based on basic particle filter can estimate the traffic state parameters of expressway. Particle degradation will occur in the application of particle filter algorithm. The occurrence of particle degradation not only leads to the loss of particle diversity, but also concentrates most of the calculations on some particles that contribute little to the result. In order to avoid the phenomenon of particle degradation, the immune particle swarm optimization algorithm is used to optimize the particle filter, and the improved particle filter algorithm is used to verify it. The results show that the improved particle filter algorithm is compared with the basic particle filter algorithm. The improved particle filter algorithm can achieve more accurate estimation of traffic state parameters. The traffic state itself has certain fuzziness and uncertainty, which is suitable to be divided by fuzzy theory. In this study, the traffic state is divided into three types by fuzzy C-means clustering method: smooth flow, mild congestion and congestion. According to the estimation results of traffic state parameters, the traffic state estimation results are obtained by clustering them into the corresponding states. On the basis of the accurate estimation of the state of the expressway traffic, the main line of the expressway and its ramp can be effectively controlled so as to maximize the service capacity of the expressway. By combining variable speed limit control with on-ramp control and considering the characteristics of different traffic conditions, a joint expressway control model is established. Genetic algorithm is used to solve the model, and an example is given to verify the model and its application. The results show that the joint control model can effectively control the expressway.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491

【参考文献】