转向延误和路段容量双约束下的用户均衡模型及算法研究  

Research on User Equilibrium Model and Algorithm with Turning Delays and Link Capacity

作　　者：卢顺达 程琳[2] 邵娟 

机构地区：[1]深圳市城市交通规划设计研究中心有限公司,深圳518021 [2]东南大学交通学院,南京210018 [3]北京交通工程学会,北京100071

出　　处：《交通工程》2017年第6期19-24,共6页Journal of Transportation Engineering

摘　　要：为了更真实地反映拥堵交通网络实际情况,考虑到交叉口转向延误和路段容量约束对交通分配模型的影响,首先构建了带转向延误和路段容量双约束下的用户均衡模型,利用Karush-Kuhn-Tucker条件证明其等价性;其次,通过改进的最短路算法,利用Projection Gradient算法作为内循环、增强拉格朗日乘子法作为外循环求解该模型.最后,利用算例路网证明该算法的可行性,对比初始状态、管控措施及双约束状态下的均衡网络流交通分配结果,发现得到路径解的同时能清晰地知道某一路段或路径交通流量转移情况,该模型通过流量转移为交通管控措施提供决策依据,进而有效地缓解交通拥堵.In order to more accurately reflect the actual situation of congested traffic network,this paper considers the influence of turning delays and link capacity constraint on traffic assignment model, and expresses its connotation in mathematical language. Based on this, the user equilibrium model with turning delays and link capacity is constructed first; the Karush-Kuhn-Tucker condition is employed to achieve the equivalence proof. Secondly, through the improved shortest path algorithm, the Projection Gradient algorithm is used as the inner loop,and the Lagrangian multiplier method is used as the outer loop to solve the model. Finally, this paper uses the example network to prove the feasibility of the algorithm,which compared with the balanced distribution of the network under the initial state, the control measures and the double constraint state. It is found that the algorithm can get not only the path solution,but also a clear understanding of a link or routetraffic flow transfer situation. This model through the flow transfer can effectively alleviate traffic congestion.

关 键 词：转向延误 路段容量 用户均衡模型 斜梯度牛顿算法(PG算法) 增强拉格朗日乘子法 

分 类 号：U121[交通运输工程]