The Study On Coordinate Fuzzy Control And Optimization Algorithms About Signal Of Urban Adjacent Traffic Intersections

With the development of economy and improvement of technology, transportation has become an essential part of economic life. However, the traditional mode of traffic control has not met the needs of the current urban transportation. As the complexity of urban traffic and the limitation of traditional traffic control method, the thesis applies the fuzzy control theory and genetic algorithm to the research area of intersection signal control in order to decrease the queue length in intersection as short as possible. So it is significant to study the optimizing timing plan.Based on the status analysis of correlational research on urban traffic control both in domestic and overseas field, it is aware that the single intersection theoretical studies have been relatively mature. But the effect of enhancing traffic capacity of the whole traffic network by using control of single intersection is still unsatisfied. The fundamental approach to work out the problem of urban transport is studying on Multi-intersection. However, there are still lots of shortages. The thesis mainly focuses on cooperative fuzzy control of adjacent traffic intersections and researches on this fact in depth.Based on the elaboration of the urban traffic control and fuzzy control theories,a multiphase classical FC which targets on decreasing the average length of queues is introduced, and it has been proven to be a more effective way than the traditional way through MATLAB simulation.The coordination control method of urban arterial traffic signal is analyzed in this thesis, and there are three optimization variables for Multi-intersection:phase difference, cycle length and green split. The first two variables illustrate the relationship among junctions, while timing optimization in single intersection is demonstrated by green split. A cooperative fuzzy control system of adjacent traffic intersections is introduced in this thesis. This system comprises two functional layers. The first layer is coordinate unit of signal system which provides optimization cycle length and phase difference for intersection based on theory of green wave. The second layer is the intersection control unit, which optimizes the signal timing through analyzing the length of queues in individual intersections. After analysing the signal timing systems, the two-stage and two-modify FC algorithm, which on the basis of traditional FC algorithm and classic two-stage FC algorithm is introduced in this thesis to control the green light delay in phases and phase switching. The two-stage and two-modify FC algorithm has been proved to be able to shorten the length of queues.Because the membership function of initial set FC may not always fit the requirement of traffic flow. The thesis adopts adaptive genetic algorithm to optimize membership function of FC on the basis of past data and experience record. This method can make the fuzzy membership function adapt to alteration in different traffic conditions. Through MATLAB simulation, it is confirmed that the optimized FC is more effective than the non-optimized one in control effect.