Research On Coordinated Control Method Of Urban Intersection Group

The continuous increase of vehicle ownership and the accelerating urbanization process have led to the intensification of traffic congestion,making the effective control of road network intersections become a research hotspot in the field o f intelligent transportation.In practical applications,the control of the intersection group is not only affected by factors such as vehicles and pedestrians,but also by factors such as road network density,road conditions and intersection signal lights.Therefore,considering various types of traffic factors,it is found that the road network traffic delay can be greatly reduced by accurately identifying the traffic state of the intersection group,subdividing the sub-areas according to the corresponding principles,and coordinated controlling of the traffic signals at each intersection in the sub-area.It can improve the overall traffic efficiency,and ultimately effectively alleviate traffic congestion.In this thesis,the coordinated control methods of urban intersection groups from three aspects: traffic status identification,associated intersection sub-area division and sub-area signal control.The specific work is as follows:1.The problem of traffic status identification of intersection groups is studied.Considering the continuity of traffic flow between road sections and intersections,a traffic state recognition method based on improved fuzzy clustering algorithm is proposed for the problem that the contribution of each index to the recognition result is different and the traffic condition recognition rate is low.This method takes the key intersection and its upstream section in the road network as a recognition unit,which determines the weight of each index through the information entropy theory to improve the objective function of the fuzzy clustering algorithm.Meanwhile,the improved algorithm is used to cluster the traffic data to improve the traffic status identification rate.2.The problem of dynamic division of associated intersections is studied.A dynamic partitioning method based on improved soft sets is proposed to deal with the problem that the partitioning factors such as intersection spacing,signal period and traffic flow fail to fully reflect the various related factors.In this method,the dynamic and static factors of intersection correlation are considered comprehensively and five partition factors are selected.Firstly,the partition factor set and its corresponding weight are taken as the input of the soft set.Then,the coordination coefficient of the adjacent intersection is taken as soft set output.Finally,according to the output value of the size of the realization of the reasonable division of the association crosshair area.3.The problem of signal coordination control at the intersection of sub-areas is studied.In order to solve the problem that the adjacent intersection are related and the premature convergence of genetic algorithm,a sub-region signal optimization method based on improved genetic algorithm is proposed.The method uses the intersection with strong correlation as a signal control unit.On the one hand,the shared function of the niche idea is introduced to adjust the group fitness.On the other hand,the crossover probability and mutation probability of the genetic algorithm are adaptively adjusted.In the end,the improved algorithm is used to optimize the average delay time to achieve effective control of the intersection.