基于遗传算法的动态模糊神经网络城市快速路入口匝道控制  

Dynamic Fuzzy Neural Network for Controlling Entrance Ramp of Urban Expressway Based on Genetic Algorithm

作　　者：张伟[1,2] 肖日东 邓晶 ZHANG Wei XIAO Ri-dong DENG Jing(School of Information Engineering, Chang＇an University, Xi＇an Shaanxi 710064, China China Highway Engineering Consulting Corporation, Beijing 100083, China)

机构地区：[1]长安大学信息工程学院,陕西西安710064 [2]中国公路工程咨询集团有限公司,北京100083

出　　处：《公路交通科技》2017年第2期129-134,148,共7页Journal of Highway and Transportation Research and Development

摘　　要：为了实现城市快速路的入口匝道智能动态控制,通过建立入口匝道数学模型,并应用具有递归环节的动态模糊神经网络于匝道控制系统中。在模糊神经网络第二层中加入内部反馈连接,使控制系统更好地响应复杂多变的交通状况,解决了以往静态网络无法处理的暂态问题。控制入口匝道的动态模糊神经网络使用遗传算法与反向传播BP算法相结合来训练,遗传算法的宏观搜索能力及鲁棒性强等优点有效地避免了神经网络算法易陷入局部极小及震荡效应等缺点。通过仿真结果,验证了基于动态模糊神经网络的控制算法相对于经典的ALINEA入口匝道控制算法具有改善,能够更好地保证城市快速路的通行效率。In order to realize the intelligent dynamic control for entrance ramp of urban expressway,the mathematical model o entrance ramp is established,and the dynamic fuzzy neural network with recurrent link is applied to ramp control system. By adding internal feedback connections on the second layer of fuzzy neural network,the control system can react complex and volatile traffic situations,the transient problem in static network which cannot be handled the past can be resolved. The combination of genetic algorithm and back propagation BP algorithm is used to train the dynamic fuzzy neural network of entrance ramp control,therefore the advantages of macro search and robustness of genetic algorithm can effectively avoid the shortcomings of neural network algorithm easily trapping into local minima and shock effects. The simulation result verifies that the control algorithm based on dynamic fuzzy neural network has obvious advantage compared with classical ALINEA entrance ramp control algorithm, especially in terms of better improving the traffic efficiency of urban expressway.

关 键 词：交通工程 城市快速路 动态模糊神经网络 入口匝道控制 智能交通系统 

分 类 号：U491.1[交通运输工程—交通运输规划与管理]