Research On Adaptive Signal Light Timing For Regional Traffic Optimization

With the acceleration of urbanization,the phenomenon of urban traffic congestion is becoming more and more serious.The timing of traffic lights oriented to regional traffic optimization becomes particularly important.At present,there are still various defects in relevant studies,such as providing static optimization schemes,optimizing only for single intersections,and insufficient coordinated control performance of trunk lines.Aiming at the above problems and aiming at optimizing regional traffic,this paper proposes solutions from three aspects: real-time performance,overall coordination and regional trunk traffic efficiency of regional traffic signal control.The main contributions of this paper are as follows:First,in order to improve the real-time performance of regional traffic signal control,a multiintersection traffic flow prediction model CNNformer is proposed to predict the intersection inlet traffic flow.The model uses convolutional neural network and Transformer to extract the spatial and temporal characteristics of traffic flow,and also considers the periodic trend of traffic flow.The average pooling layer is used to extract the periodic characteristics of traffic flow.Experiments show that compared with the existing multi-intersection traffic flow prediction model DISTN,CNNformer can improve the prediction accuracy by 10.2%.Secondly,in order to improve the overall coordination of regional traffic signal control,a dynamic regional traffic signal coordination optimization control method is proposed.This method is based on the predicted intersection inlet flow and combined with the improved whale optimization algorithm proposed in this paper to obtain the optimal solution of the regional vehicle average delay model,so as to realize the dynamic optimization of regional traffic signal timing.By improving the population initialization method and individual iteration updating method of whale optimization algorithm,the local development and global exploration ability can be improved.The comparison experiment shows that compared with the whale optimization algorithm,the improved whale optimization algorithm has stronger searching ability and can reduce the average delay time of regional vehicles by 8.1%.The simulation results show that compared with the fixed timing control method,the dynamic regional traffic signal coordination optimization control method can improve the regional road network traffic efficiency by 21.9%.Finally,in order to improve the traffic efficiency of regional trunk lines,a bidirectional green wave adaptive coordination control method is proposed.This method is based on the optimized results of dynamic regional traffic signal timing and the multi-strategy fusion graphic method proposed in this paper.By adjusting the combination form of signal phase and green time of each intersection on the trunk,the multi-strategy fusion diagram improves the flexibility of phase timing and can better match the arrival time of two-way traffic on the trunk.Compared with the traditional graphic method,the multi-strategy fusion graphic method can improve the bandwidth of green wave by about 20%.The simulation results show that compared with the fixed timing control method,the two-way green wave adaptive coordinated control method can improve the traffic efficiency of the trunk line by 32.3%,and compared with the dynamic regional traffic signal coordinated optimization control method,it can improve the traffic efficiency of the regional road network by 8.7%.