基于粒子群算法的多匝道协调控制

Coordinated Multi-Ramp Control Based on Particle Swarm Algorithm

摘要:

高速公路入口多匝道协调控制是调节各个入口匝道进入到主线的交通量,从而使主线交通流处于最佳状态。由于多个入口匝道相互关联和相互影响,多匝道协调控制具有强的耦合性、非线性和时变性。针对上述问题,提出了一种基于系统分层和粒子群优化算法的控制方法。首先论述了高速公路多匝道系统的原理模型；然后阐述了多匝道协调控制系统的实现,系统由协调控制层和直接控制层组成,其中前者负责模型选择、参数调整和确定期望的密度轨迹,后者采用比例积分微分控制器实施控制,并引入粒子群算法对控制器的比例系数、积分系数和微分系数进行优化；最后进行了仿真实验。结果表明,当高速公路出现交通拥堵时,系统能够快速地消除拥堵,并使主线交通流趋于稳定。该方法为高速公路多匝道控制提供了一种切实可行的新途径。

Abstract:

Coordinated multi-ramp control at freeway entrance is to regulate the traffic volume entering into freeway mainlines from all on-ramps. Thus, mainline traffic flow is at its best. The coordinated multi-ramp control system is a strong coupling, nonlinear, and time-varying system because of the interaction of multi-ramps. Aiming at the problems as mentioned above, a coordinated control method is proposed based on a system hierarchy and a particle swarm optimization algorithm. Firstly, the principle and model of freeway multi-ramp system are dissertated. Then, the realization of a coordinated multi-ramp control system is expounded. This control system is composed of a coordination control layer and a direct control layer. Coordination control layer is used to select traffic models, to adjust the model parameters, and to determine desired trajectories of traffic densities. The direct control layer is responsible for control implementation by using proportional-integral-derivative controllers. A particle swarm algorithm is utilized for optimizing the proportional coefficients, integral coefficients, and derivative coefficients. Finally, simulation experiments are carried out. Simulation results show that the control system can quickly eliminate traffic jams and can stabilize the mainline traffic flow when freeway traffic congestion occurs. This approach provides a new and practical way for freeway coordinated multi-ramp control.