高速路入口匝道的模糊神经网络自适应协调控制  

Adaptive coordinated control based on FNN for freeway entering ramp

作　　者：徐进学[1] 于双和[1] 徐丽君[2] 

机构地区：[1]大连海事大学信息科学技术学院,大连116026 [2]沈阳工业大学信息科学与工程学院,沈阳110023

出　　处：《东南大学学报（自然科学版）》2009年第S1期266-271,共6页Journal of Southeast University：Natural Science Edition

基　　金：国家自然科学基金资助项目(60674061)

摘　　要：为了实现高速公路多路段入口匝道的协调控制,提出了一种基于模糊神经网络的自适应协调控制方法.该方法首先利用模糊规则实现了相邻路段间交通状态的协调,并对匝道排队进行调节,使其不超过最大排队长度.然后,在神经网络权值优化过程中,采用遗传算法对隶属度函数进行优化,避免算法收敛于局部最优解.该方法具有不依赖于系统的精确模型、控制算法能自适应外界变化、可实现多路段间协调控制等优点.仿真结果表明,该方法对3个路段的高速公路入口匝道能够较好地实现自适应协调控制;与经典的ALINEA方法相比,优化速度更快,在抑制交通密度波动和排队长度增长方面效果更好,对道路容量的利用也更加充分.To implement the coordinated control for multi-section entering ramps of freeway,an adaptive coordinated control method is presented based on fuzzy neuron network(FNN).First,by using fuzzy rules,the coordination of the traffic states for adjacent sections is realized and the ramp queue length is regulated to avoid exceeding the maximum.Then,the genetic algorithm is adopted for the optimization of FNN in order to avoid converging to a local optimized solution.The proposed method is not dependent on the exact model of system,and can adapt the changes in the outside world and achieve the multi-section coordinated control.The simulation results demonstrate the effectiveness of the method for 3-section entering ramps.Compared with the classic control method,ALINEA,the optimized speed of this method is faster,the effect in the aspects of suppression of traffic density fluctuation and queue length growth is better,and the fuller use of road capacity is available.

关 键 词：模糊神经元网络 高速公路入口匝道 自适应协调控制 

分 类 号：U491.54[交通运输工程—交通运输规划与管理]