中文摘要：

针对高速公路匝道控制中单一性能指标评价具有片面性的问题,提出了一个基于多性能指标优化的高速公路匝道控制方法。在应用METANET模型进行交通流特性与控制建模的基础上,设计了主线运营效益、排队长度和排队公平指数3个性能指标,用于描述高速公路路网的运营效率和公平性。根据Pareto原理,利用非支配性快速遗传算法对多个性能指标进行综合优化,依据所求得的较优解集筛选最优解。以宁杭高速溧水—江宁段的匝道控制为例,从Pareto最优解筛选、排队长度的变化状况、控制效果比选等方面进行了比较。分析结果表明：该方法可综合考虑多个指标,减少对工作经验的依赖,使匝道控制效果在效率和公平的博弈中达到最好。

英文摘要：

Aimed at the problem of unilateral evaluation of single-performance index on expressway ramp metering,a control method of multi-performance indexes optimizing was proposed.Based on the modeling of traffic flow characteristics and control system by using METANET model,main-stream traffic efficiency,queue length and queue fairness index were designed to describe the efficiency and fairness of expressway network.According to Pareto principle,non-dominated sorting genetic algorithm Ⅱ（NSGA-Ⅱ） was introduced to optimize several performance indexes at the same time,so that the optimal solution could be found from Pareto solutions.Taking Lishui-Jiangning section of Ninghang expressway as an example,optimal solution from Pareto solutions,variation of queue length and control effect were analyzed.The results indicate that this method can consider several performance indexes comprehensively,reduce the dependence on working experience and improve the metering efficiency and fairness.