论文部分内容阅读
利用递阶结构和模糊神经网络来进行交通系统的实时协调控制。其基本思想是把交通干线作为一个大系统。子系统为干线上的各交叉口,用模糊神经网络综合调控绿信比,相位差以及周期时长等3方面因素,在模糊神经网络控制器的设计中使用4层网络,将3种数据模糊化后输入,进而得到相应的输出结果,目的在于更好的协调交叉口间的信号,使得主干路的排队长度最小,从而减少车辆延误。最后将设计的控制器作用于北京市海淀区学院路与成府路和清华东路所构成的系统进行仿真研究,结果显示,该方法在减少主干道车辆延误的同时,综合考虑了次干道路段的需求,使得在减少车辆延误方面有较为显著的效果。
The use of hierarchical structure and fuzzy neural network for real-time coordination of traffic control system. The basic idea is to use the traffic trunk as a large system. The subsystem is the intersection on the main line, and uses the fuzzy neural network to synthetically control the green signal-to-noise ratio, the phase difference and the period duration, and uses the 4-layer network in the design of the fuzzy neural network controller to blur the 3 kinds of data After input, and then get the corresponding output results, the purpose is to better coordinate the signal between the intersection, making the trunk line queue length minimum, thereby reducing vehicle delays. Finally, the controller is designed to simulate the system formed by Xueyuan Road, Chengfu Road and Qinghua Road, Haidian District, Beijing. The results show that this method not only reduces the delay of main road vehicles, but also considers the secondary road sections The demand has made, in the reduction of vehicle delays have a more significant effect.