基于粗糙集高速公路混沌T-S FNN控制仿真

Chaos Control Simulation of Freeway by Using T-S Fuzzy-neural Networks Based on Rough Sets

研究基于粗糙集理论的高速公路混沌系统模糊神经网络入口匝道控制方法。针对高速公路车流量不确定性特点,提出了通过数据挖掘技术建立交通流入口匝道智能混沌控制器知识库的思想;设计了以密度、上游流量和最大李亚普诺夫指数作为输入,红灯时间作为输出的T-S模糊神经网络混沌控制器;采用粗糙集理论建立混沌控制器知识库,确定模糊神经网络控制器结构并提取模糊规则;采用模糊神经网络方法对控制器参数进行优化。仿真结果表明:采用该方法设计的智能混沌控制器,可实现保持高速公路有序运动、避免交通堵塞、提高交通通行能力的目的,是提高高速公路管理控制水平的有效方法。

The on-ramp chaos control problem of freeway was studied by using fuzzy-neural networks based on rough set theory. Based on the uncertainty characteristic of vehicle volume in freeway system, the thought was proposed to establish the knowledge base of the on-ramp intelligent chaos controller of freeway by using data mining technology. The T-S fuzzy- neural networks ramp controller was designed. The traffic density, upstream traffic volume and maximal Lyapunov exponents are the input variables and the red timing is the output variable of the controller The knowledge base of the chaos controller was established by using rough sets theory. The controller structure including the extracting of fuzzy rules was determined. The parameters of the fuzzy controller were optimized by using fuzzy-neural networks. The feasibility and validity of the method was demonstrated via the simulation experiment, which shows that the order motion can be realized, the traffic jam phenomena can be suppressed, and the freeway capacity can be enhaned by using the intelligent chaos controller.