交叉口定周期信号控制多目标优化方法  

Multi-objective Optimization Method for Fixed-time Signal Control at Intersection

作　　者：高云峰[1] 徐立鸿[2] 胡华[3] 韩皓[1] 杨晓光[4] 

机构地区：[1]上海海事大学交通运输学院,上海200135 [2]同济大学电子与信息工程学院,上海201804 [3]上海工程技术大学城市轨道交通学院,上海201620 [4]同济大学交通运输工程学院,上海201804

出　　处：《中国公路学报》2011年第5期82-88,共7页China Journal of Highway and Transport

基　　金：国家自然科学基金项目(51008231);中国博士后科学基金项目(20090460653;201104272);上海市博士后科研资助计划项目(10R21416500);上海市重点学科建设项目(S30601)

摘　　要：为了克服以往采用加权组合方法将多目标优化模型转为单目标优化模型进行求解的弊端,以非饱和交叉口为研究对象建立了定周期信号控制参数的多目标优化模型,并以非支配排序遗传算法(Non-dominated Sorting Genetic Algorithm II,NSGA II)为基础建立了直接求解单个交叉口交通信号多目标控制模型的相容优化方法。最后通过实例分析了延误、停车率、排队长度等常用交通信号控制目标的有效性和对控制参数的敏感性。研究结果表明:与单目标优化方法相比,多目标相容优化方法更能减少车辆在交叉口上的停车延误和排队长度;与加权组合方法相比,多目标相容优化方法能获得更优的综合交通效益。In order to overcome the solution deficiencies which was used to transform multi-objective optimization model, a multi objective optimization model of fixed time of the weighted combination method model into single objective optimization signal control parameter of unsaturated intersection was set up, and based on non-dominated sorting genetic algorithm Ⅱ （NSGA Ⅱ）, a multi-objective compatible optimization algorithm which could be used to solve multi-objective optimization problems of single intersection directly was established. Validities of common control objectives, such as average vehicle delay, average vehicle stops, and average vehicle queue length and sensitivities of the objectives to signal control parameters were analyzed by example. Results show that compared with the single objective optimization method, the multi-objective compatible optimization method can decrease average vehicle delay and queue length at intersection. Compared with the weighted combination method, the muhi-objective compatible optimization method can obtain better comprehensive traffic efficiencies.

关 键 词：交通工程 交叉口 多目标相容优化 信号控制 有效性 敏感性 

分 类 号：U491.5[交通运输工程—交通运输规划与管理]