Research Of Ramp Metering Strategy Of The Traffic Corridor System: An Continuous Approximation Method

The transportation industry is shifting from pure traffic supply by scale expansion to refined management focusing on innovation-driven.Meanwhile,with the remarkable growth of the freeway and arterial network,ramps,as the connecting section of the two,are becoming more and more densely distributed.Unfortunately,the disorderly influx of vehicles or overcontrol of ramps will induce congestion not only on freeways but also on surface roads.Thus,it is important to apply ramp metering to curb traffic congestion.Effective ramp metering can coordinate the flow layout of the freeway and arterial network,and optimize the operation of the transportation corridor network(including freeway mainlines,ramps,and surface roads).To this end,this paper proposes a bi-level game-theoretical model to optimize the system performance of a corridor network and consider the interplay between System Optimal(SO)and User Equilibrium(UE).In the upper model,from the perspective of the traffic authority,the ramp metering strategy is formulated to minimize the total travel time of the corridor network.In the lower model,travelers choose the route with the shortest time in response to the designed ramp metering scheme.With the designed ramp metering scheme,the bi-level model comprehensively considers the mutual interaction between the traffic authority and travelers,and finally achieves the optimal system performance.To simplify and accelerate the solving process of the proposed model,the third chapter introduces Continuous Approximation(CA)to explore the travel behavior under UE:When the travel distance is no greater than the distance splitting point κ,travelers will only choose surface roads,otherwise,travelers will only choose freeway.By clarifying the traffic diversion behavior,the original bi-level model can be transformed into a single-level optimization model with regard to κ.The simplified single-level model is solved by an analytical algorithm and a numerical algorithm guided by CA.Firstly,the analytical algorithm developed in the fourth chapter discusses the solving process of the proposed ramp metering model.Based on the analyticalform solution,the active condition for ramp metering is explored.Then,the accuracy and efficiency of the numerical algorithm are testified by comprehensive experiments in the fifth chapter.In the sixth chapter,the optimization effects are proved that the developed ramp metering strategy and traffic diversion behavior can optimize the total travel time of the transportation corridor system.Furthermore,the parameter analysis explores the influence of traffic demand(background traffic and travel demand)and road network parameters(capacity and free-flow time)on the traffic diversion behavior.Results show that(1)The ramp metering strategy designed by this research can reduce the total travel time of the traffic corridor system by 10.69%.(2)The traffic diversion information can reduce the total travel time of the traffic corridor system by 8.84%.(3)The background traffic is the most important determinant of the change trend of κ,which determines the moving trend of κ when travel demand changes.(4)Freeway capacity is the most important parameter that causes numerical changes of κ.When a 5%fluctuation is applied to the freeway capacity,the resulting change rate of κ is 5.6%.(5)The change of ramp free-flow time has basically no effect on the distance splitting threshold κ.