Optimizing Traffic Signal Settings in Smart Cities
Optimizing Traffic Signal Settings in Smart Cities
复制标题
DOI：
10.1109/tsg.2016.2526032
复制
发表时间：
2017-09
影响因子：
9.6
通讯作者：
Zhiyi Li;M. Shahidehpour;S. Bahramirad;A. Khodaei
中科院分区：
文献类型：
--
作者：
Zhiyi Li;M. Shahidehpour;S. Bahramirad;A. Khodaei
交通信号在智慧城市中发挥着关键作用，可缓解大都市地区的交通拥堵和减少排放。本文提出了一种双层优化框架来解决最优交通信号设置问题。上层问题确定交通信号设置以最小化驾驶员的平均出行时间，而下层问题旨在使用上层计算的设置实现网络平衡。采用遗传算法与基于微观交通模拟的动态交通分配（DTA）相结合，将复杂的双层问题解耦为易于处理的单层问题，并依次求解。对合成交通网络和现实交通子网进行案例研究，以检验所提出模型和相关解决方法的有效性。为扩展所提出的模型和加速大面积交通网络应用中的求解过程提供了额外的策略。
Traffic signals play a critical role in smart cities for mitigating traffic congestions and reducing the emission in metropolitan areas. This paper proposes a bi-level optimization framework to settle the optimal traffic signal setting problem. The upper-level problem determines the traffic signal settings to minimize the drivers’ average travel time, while the lower-level problem aims for achieving the network equilibrium using the settings calculated at the upper level. Genetic algorithm is employed with the integration of microscopic-traffic-simulation-based dynamic traffic assignment (DTA) to decouple the complex bi-level problem into tractable single-level problems, which are solved sequentially. Case studies on a synthetic traffic network and a real-world traffic subnetwork are conducted to examine the effectiveness of the proposed model and relevant solution methods. Additional strategies are provided for the extension of the proposed model and the acceleration of solution process in large-area traffic network applications.