考虑汽车排放需求变化的多目标连续型交通网络设计  

Multiple objective analysis of continuous transportation network design under composite demand

作　　者：柴泾哲 彭建刚[3] 石琴[1] 孙飞[1] 

机构地区：[1]合肥工业大学交通运输工程学院,安徽合肥230009 [2]城市交通管理集成与优化技术公安部重点实验室,安徽合肥230009 [3]合肥工业大学汽车工程技术研究院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2015年第12期1591-1596,共6页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(71431003);国家科技支撑计划资助项目(2014BAG03B02);安徽省软科学研究计划重点资助项目(1302053022)

摘　　要：文章研究需求变化的多目标连续型交通网络设计问题的优化模型和算法,利用双层规划模型求解问题;上层以路网系统阻抗、路段总投资、汽车不同尾气排放量最小化作为优化目标,并受到原路段的通行能力约束,下层是基于需求变化下的用户平衡配流模型;使用非对称Nguyen-Dupuis网络,利用精英保留和随机遍历的选择遗传算法求解上层模型,采用基于路径的双梯度投影算法求解下层模型。设计相应算法程序对模型进行验证,通过算例测算综合需求排放分析和综合需求下多目标参数分析,结果表明模型有效、求解算法可行。In this paper, the optimal model and algorithm of multiple objective continuous transportation network design problem under composite demand are presented. A bi-level programming model is used to solve the problem. The upper model considers the minimization of network impedance, total investment and vehicle exhaust emissions as the goal and is subjected to the traffic capacity constraints of original section. The lower model is the user equilibrium assignment model under different demand. The asymmetric Nguyen-Dupuis neural network and the selection genetic algorithm with elitist strategy and random traversal are used to solve the upper model, while the lower model is solved by the dual gradient projection algorithm based on path. The corresponding algorithm is designed to verify the model, and the results of the example analysis show that the model is effective and the algorithm is feasible.

关 键 词：双层规划模型 遗传算法 双梯度投影算法 尾气排放 

分 类 号：U491[交通运输工程—交通运输规划与管理]