Congestion Pricing Model And Algorithm Based On Stochastic User Equilibrium

With the continuous deepening of economic development and the rapid progress of urbanization process,under the pace of rapid urban development,the city is expanding outward and the economic development regionalization,resulting in the low efficiency of urban traffic network operation,which makes the urban traffic congestion problem remarkable,And traffic congestion charges,which is an effective means for adjusting urban congestion road sections or congested areas,adjust the traffic flow through charging a certain fee for the vehicle.the traffic demand of OD is allocated to each path,reducing the traffic load and alleviating the traffic congestion.Based on the elastic demand transportation network and the fixed demand transportation network,two congestion pricing bilevel programming model is established for stochastic user equilibrium.For the congestion pricing model under the elastic demand in the network,the upper level programming problem is to maximize the social and economic benefits of the road network traffic.The lower level model is the stochastic user equilibrium problem.In the solution algorithm,the model is solved by combining modified truncated Newton algorithm and annealing simulation algorithm.For the congestion pricing model under fixed demand,the upper level programming problem is to minimize the total travel time of the system,and the lower level model is a stochastic user equilibrium problem.The model is solved by combining genetic algorithm and quasi-Frank-Wolfe algorithm.Then a numerical example is used to compare the traffic flow and economic benefits before and after the charge.Finally,the corresponding conclusions are drawn to verify the validity of the model and the effectiveness of the algorithm.