Dissertation > Excellent graduate degree dissertation topics show

The Study on Urban Traffic Signal Control Method at Single Intersection

Author: BaiLiFang
Tutor: XuJinXue
School: Dalian Maritime University
Course: Control Theory and Control Engineering
Keywords: Fuzzy Control Genetic Algorithms Neural - Fuzzy Control System Modeling System Simulation
CLC: U491.51
Type: Master's thesis
Year: 2011
Downloads: 111
Quote: 1
Read: Download Dissertation

Abstract

Rapid economic development and urbanization process accelerated exacerbated urban traffic congestion. And bring vehicle delay increases, frequent accidents and environmental degradation series of problems. Intelligent traffic control system is an important way to solve this problem. For urban traffic control systems are random and highly nonlinear problems such as difficult to establish accurate mathematical model of this shortcoming, this paper takes advantage of fuzzy logic, genetic algorithms and neural networks intelligences theory, a study of single intersection to ensure that traffic can smoothly pass the minimum delay to ease urban traffic congestion resulting harm. General signal optimization programs are targeted under normal circumstances, and for the lane of traffic accidents or driveway repair does not consider the situation, this paper designs a suitable both for normal and non-normal case signal control scheme. First, from the perspective of cybernetics fuzzy control method for controlling a Single Intersection cities fuzzy control model, on this basis, the application of fuzzy control theory and the experience of the traffic police directing traffic, is designed based on a single vehicle delay intersection traffic signal control strategy. The policy objectives of the vehicle as a control delay, considering the average vehicle adjacent to the phase delay and the current lane situation, in order to determine when each phase of the green allocation. Second, for the classic fuzzy controller defect designed fuzzy controller based on genetic algorithm optimization. Optimization using genetic algorithms to adjust the amount of fuzzy control rule rather than the control rules themselves, to avoid the unreasonable rules, and to improve the convergence speed. Again, for the neural - fuzzy control method, this paper introduces adaptive neuro-fuzzy inference system structure principle and reasoning methods, and the use of self-learning function neural networks, fuzzy inference system (FIS) in the fuzzy logic rules and membership functions parameters through neural network learning to self-tuning, automatically generate fuzzy rules and adjust the membership function to solve the fuzzy control system of fuzzy inference rules are mainly based on traffic police experience in design, lack of self-learning ability, not high control accuracy. Finally, the use of Matlab and Simulink tools were the results of the three control methods for modeling and simulation. Simulation results show that the three control strategies for normal and non-normal conditions can effectively reduce vehicle delay, the effect is good, and after optimization of the control is better than traditional fuzzy control. In summary, this paper combines fuzzy logic, genetic algorithms and neural networks based on the theory of artificial intelligence, in-depth study of urban traffic signal control strategies, effectively reducing the average vehicle delay.

Related Dissertations

CLC: > Transportation > Road transport > Technical management of traffic engineering and road transport > Traffic engineering and traffic management > Line of traffic safety facilities > Traffic signals
© 2012 www.DissertationTopic.Net Mobile