| Urban Road intersection is an important part of urban road network,and its traffic efficiency affects the capacity of the whole road network.Multi period signal control at intersection is an important means to improve traffic efficiency and reduce vehicle delay.At present,the signal timing optimization based on time division at intersection is mostly based on the total flow of the whole entrance road,which is inconsistent with the actual traffic flow of phase control,and there are many discontinuous time periods,It affects the actual control effect of intersection traffic flow.Based on this,this paper studies the optimization of intersection signal timing based on multi period division from the point of view of traffic flow data similarity,timing and phase control.Firstly,considering the similarity and time sequence of intersection traffic flow data,two multi time division methods are proposed.One is the commonly used K-means clustering algorithm.Based on the analysis of the law and similarity between the traffic volume data,the gap between the traffic flow sample data is taken as the similarity evaluation index to measure the similarity between the samples,and the continuous traffic flow data is divided into several time segments according to the similarity value.The other is an ordered clustering algorithm based on traffic flow time sequence,and then considering the time sequence of traffic flow data,aiming at the situation that there is discontinuity between the divided time periods,with the help of time sequence segmentation algorithm,an algorithm based on traffic flow data is designed to minimize the sum of inner diameter,and a dynamic search algorithm is designed to solve the problem.The divided time periods are used to match different signal periods and timing schemes.Secondly,on the basis of considering the actual traffic flow controlled by the signal timing phase,a signal timing method for intersection based on vehicle delay index optimization is proposed to match different timing schemes for each time period after traffic control division.In the selection of parameters,the cycle length and the green signal ratio are optimized.The phase and eight different traffic flow directions are combined as the design parameters of the total traffic delay algorithm,and the traffic light length is allocated according to the traffic flow dynamically converted in each phase of the intersection signal.On this basis,taking the saturation and cycle length of each phase as the constraints,a signal phase timing method based on time division is proposed and solved by genetic algorithm.Finally,taking the actual intersection as an example,the effectiveness of the optimization method is verified according to the traffic flow data.The simulation results by VISSIM show that,compared with the original signal control scheme,the control effect is more accurate and effective,and the total vehicle delay at the intersection is effectively reduced by 9.1%,which shows that the proposed method has a certain practicality. |
