Study On Intelligent Chaos Control Of Traffic Flow

The existing traffic control methods are presented based on the traditional theory and the general control principle, and are not related to the chaos of traffic flow. Therefore, the study on the chaos control of traffic flow is an important new problem. Some main aspects, which were studied in this dissertation, are presented as follows:The real-time rapid recognition problem of chaos in traffic flow was studied by using support vector machine. Based on analyzing the demand of traffic control to chaos recognition of traffic flow and the problems of the exiting chaos recognition methods, the rapid real-time recognition method of chaos in traffic flow was brought forward by using support vector machine. The principle and the structure of the system are briefly introduced. The extracting of the feature vector and the two algorithm of online recognition in support vector machine model are discussed mainly. The simulation result shows that the method is correct and feasible.The principle of chaos control in freeway system was studied. Based on the object description of chaos control in a freeway section with only one entrance ramp, the principle was proposed and the red timing was selected as the control variable.The problem of chaos control in freeway system was studied by using delay feedback fuzzy chaos control. The synthetic method was proposed to control the chaos system of freeway by using the theory of delay feedback control and the fuzzy control method. The feedback ramp controller was designed. The goal of chaos control is to realize periodic orbits from unstable periodic movement and the delay error of traffic density and the variable of the delay error are the input variables of the controller. The Mamdani fuzzy controller and the T-S fuzzy controller were discussed respectively. The parameters of the fuzzy controller were optimized by using genetic algorithm. The simulation result shows that the method is correct and feasible.The problem of chaos control in freeway system was studied by using state delay feedback fuzzy chaos control. The synthetic method was proposed to control the chaos system of freeway by using the theory of delay feedback control and the fuzzy control method. The state fuzzy ramp controller was designed. The traffic density and upstream traffic volume are the input variables of the controller. The Mamdani fuzzy controller and the T-S fuzzy control were discussed respectively. The parameters of the fuzzy controller were optimized by using genetic algorithm. The simulation result shows that the method is correct and feasible.The problem of chaos control in freeway system was studied by using fuzzy-neural networks method based on subtractive clustering. The thought was proposed to establish the knowledge base of the chaos controller by using data mining technology. The fuzzy T-S ramp controller was designed. The traffic density, upstream traffic volume and maximal Lyapunov exponents are the input variables of the controller. Subtractive clustering is used to determine the controller structure such as the extracting of fuzzy rules and generating initial parmeters. The parameters of the fuzzy controller were optimized by using fuzzy-neural networks. The radius of the cluster centers were synthetically optimized by genetic algorithm. The simulation result shows that the method is correct and feasible.