Method Research Of Signal Optimization And Guidance Control For Traffic Incidents

With the completion of a well-off society in an all-round way,the improvement of people’s living standards and the continuous increase of motor vehicle ownership,the speed of urban road construction has been unable to meet the traffic needs of residents.At present,the frequent traffic incidents on urban roads often lead to traffic chaos and road congestion.If not handled in time,it will lead to traffic paralysis of regional road network,which will have a significant impact on social economy and safety.Therefore,the study of signal and guidance under traffic events is of great significance to reduce traffic congestion,speed up evacuation and restore the normal operation of urban traffic.Therefore,the signal and guidance research under traffic incident conditions is of great significance to reduce traffic congestion,speed up the evacuation of incidents,and restore the normal operation of urban traffic.The main contents of this thesis are as follows:(1)Analysis of traffic flow characteristics under traffic events.Based on the traffic wave theory,the spatial-temporal characteristics of traffic flow under traffic events are analyzed to explore the internal mechanism and change rules of the evolution of traffic congestion,which provides a basis for subsequent modeling;Aiming at the impact time of traffic events,this thesis proposes a traffic event duration prediction model based on Light GBM algorithm,and trains the traffic event data of Beijing Winter Olympic Games in December 2018.The Light GBM method is compared with SVM and Random Forest prediction method,and the evaluation indexes Mae,MSE and MAPE are significantly improved,which verifies the rationality and scientificity of the prediction method.(2)Considering the different degrees of impact of traffic events,control strategies are divided into two types: upstream and downstream signal optimization and path guidance coordinated control.In view of the traffic incident signal control strategy with less impact,the traffic incident road section is segmented,which is composed of the upstream road section and the downstream road section of the incident,and the signal timing optimization model is constructed with the goal of minimizing the delay of the incident upstream and downstream intersections and road sections.Constraints on the phase difference between the upstream and downstream intersections,the coordinated control phase,and the green light duration of the non-coordinated control phase of the traffic incident.(3)In view of the signal control strategy with a large degree of traffic events,the traffic flow flowing to the event section was induced by path detour,and an optimization model of signal coordination control parameters under detour path was established.The optimization objectives of the model were to minimize the total delay of the coordination phase of the detour path and maximize the total output flow in the event area.(4)The effectiveness of the control model proposed in this thesis is verified through a case,the improved particle swarm algorithm combined with the MATLAB platform is used to solve the model,the COM interface of VISSIM is called by the PYTHON language,and the second development realizes the signal time control.The simulation results show that compared with the traditional timing scheme and the existing signal scheme,the number of passing vehicles,queue length,delay and other indicators,the model in this thesis has a better improvement effect and has certain application value.