| With urbanization and motorization,traffic congestion is more serious,and the congestion of road sections and intersections has spread to road network congestion.Some existing control models and methods require OD data,and accurate OD data is difficult to obtain,making the actual control effect poor.MFD can circumvent this problem,it is an inherent property of the road network,no specific OD data is needed,and it is often used to identify the traffic state of the road network in a short time.Therefore,based on the MFD theory,this article explores the MFD characteristics of the urban road network,thereby designing effective control strategies to alleviate road network traffic congestion.It mainly includes the following five aspects:(1)Based on Vissim,a regional road network MFD model is constructed.First,the cumulative number of vehicles and the completed flow in the regional road network are selected as MFD parameters.Then,a part of the road network in Beijing Wangjing area was drawn,the signal timing of the intersection was set,the flow input was determined through 10 times of debugging,and data collection and simulation evaluation were carried out.Then,the existence of MFD in the road network in the study area was confirmed.(2)Proposed a road network sub-region division algorithm based on Normalized Cut(NCut).Firstly,based on the similarity of traffic operation,a weight map of the road network is constructed,and the NCut algorithm flow is analyzed.Then,an evaluation index is proposed to overcome the problem that the NCut algorithm cannot optimize the number of sub-regions.Then,the proposed method is used to divide the road network into multiple homogeneous sub-regions,the MFD characteristics of each sub-region are analyzed,and the controlled area is selected from them.(3)A road network traffic state recognition algorithm based on GA-FCM is proposed.First,the characteristics of Fuzzy C-means clustering(FCM)algorithm are analyzed,and the initial clustering center is determined based on Genetic Algorithm(GA)algorithm to overcome the FCM algorithm's tendency to fall into the local optimal problem.Then,the XB coefficient is selected as the evaluation index of clustering effect,which proves that the clustering performance of GA-FCM algorithm is 31.375% better than the traditional FCM algorithm.Then,the traffic state of the controlled area was identified.(4)A boundary feedback control strategy based on fuzzy RBF neural network PID is designed.First,the regional vehicle flow balance equation is established,and a control target to maintain the cumulative number of vehicles in the region near the critical value is proposed.Then,the fuzzy RBF neural network is used to improve the PID controller,and the boundary flow is controlled.Then,through Matlab/Simulink simulation,compared with the conventional PID,the proposed method optimizes the time required to reach the optimal cumulative number of vehicles by 32.450%,and determines the final control transfer rate in each simulation cycle.(5)The second development of Vissim is based on Python,and the empirical analysis of regional boundary control is completed.First,the signal timing scheme for the border intersection was formulated,and the position of the final border control intersection was adjusted to avoid excessive queueing of vehicles at the border intersection affecting the upstream section.Then,a number of evaluation indicators are proposed,and the proposed method is compared with traditional timing signal control and fuzzy PID control.The results show that the proposed method can not only improve the traffic indicators in the controlled area,effectively alleviate regional traffic congestion,and improve the traffic operation efficiency of the entire road network,but it will not affect the traffic condition of the upstream section of the border intersection.Control means with good control effect. |
PDF Full Text Request