Research On Traffic Congestion Forecast And Control Based On Digital Twin Technology

With the increasing complexity of urban transportation systems,the state of traffic operation has become an important issue facing cities,and increasingly serious traffic congestion has brought inconvenience to travel.This paper takes the urban traffic state and traffic congestion as the research object,combined with the digital twin technology to deeply study the traffic state prediction model and traffic congestion control method.The main work and contributions are as follows:（1）Aiming at the problem of urban traffic state prediction,a feature fusion model based on improved particle swarm optimization（IPSO）optimization radial basis function（RBF）and long short-term memory（LSTM）/support vector machines（SVM）is proposed.Traffic state prediction method.This method combines traffic operation characteristics,uses feature engineering ideas to determine the factors that affect future traffic status in traffic flow data,uses the IPSO-RBF model to extract temporal and spatial characteristics of traffic flow parameters,and introduces LSTM to effectively perform the time series characteristics of traffic congestion status.Extraction,the extracted features are fused and input into the SVM to classify and predict the traffic state.Based on the collection of traffic data in some areas of Shenyang station in Shenyang city,experiments are carried out and compared with other algorithms to verify the superiority of this paper based on the feature fusion model of IPSO-RBF and LSTM/SVM.（2）Aiming at the problem of urban traffic congestion,a congested road section control method based on flow distribution is proposed.First,according to the characteristics of traffic congestion,the factors affecting traffic congestion are analyzed;then the method of road capacity judgment and vehicle safety queuing coefficient is used to calculate the remaining capacity of the road,and the traffic flow is reasonable for the road sections with congestion in the future.The allocation processing completes the effective guidance of congested road sections;Finally,VISSIM is used to simulate and compare experiments with other traffic control schemes to verify the superiority of the optimization method for congested roads in this paper.（3）In order to improve the effectiveness of urban intelligent traffic congestion prediction and control,this paper introduces digital twin technology,combining the traffic state prediction model based on IPSO-RBF and LSTM/SVM feature fusion with the congested road section control method based on flow allocation,Designed a digital twin traffic congestion prediction and control system.It mainly includes information logging system,rapid traffic modeling module,traffic prediction management module,and traffic control management module.By collecting historical data of actual traffic in some areas of Shenyang station in Shenyang city,the system is applied in a case,which verifies the applicability and feasibility of the use of digital twin technology in traffic congestion prediction and control in this paper.