进港航班排序优化数学模型研究  

Research on Optimization Mathematical Model of Arrival Flights Scheduling

作　　者：王世豪[1,2] 杨红雨[1,2] 武喜萍[3,2] 刘洪[3,2] 

机构地区：[1]四川大学空天科学与工程学院,四川成都610065 [2]四川大学国家空管自动化系统技术重点实验室,四川成都610065 [3]四川大学计算机学院,四川成都610065

出　　处：《四川大学学报（工程科学版）》2015年第6期113-120,共8页Journal of Sichuan University (Engineering Science Edition)

基　　金：国家空管科研课题资助项目(GKG201403004)

摘　　要：针对常用进港航班排序数学模型(总延迟时间最小和总延迟成本最小)中存在的问题,选取空中延误成本、旅客延误成本、后续延误成本以及环境污染成本4个指标综合建立一种改进的总延迟成本最小数学模型。在分析已有的基于模拟退火的粒子群算法(SA-PSO:particle swarm optimization based on simulated annealing)优化进港航班排序时寻优能力不足、收敛速度慢的基础上,采用一种线性微分递减(LDD:linear differential decrease)的退火策略,从而可以有效地解决进港航班排序问题。实验结果表明,与FCFS(first come first serve)、PSO以及SA-PSO算法相比,LDD-SA-PSO算法在进港航班优化问题上具有较好的寻优能力和收敛速度,同时改进数学模型中参数选择对优化结果也具有明显影响。In view of exiting shortages of the commonly used mathematical model of arrival flights scheduling,such as minimum delay time and minimum delay cost,four indicators including air delay cost,passenger delay cost,subsequent delay cost and environmental pollution cost were chosen to comprehensively establish an improved mathematical model of minimum delay cost. On the basis of analyzing insufficient ability of seeking optimization and slow convergence speed for the existing particle swarm optimization based on simulated annealing（ SA-PSO） algorithm,an annealing strategy of linear differential decrease was applied to SA-PSO（ LDD-SA-PSO） algorithm,and the arrival flights scheduling was more effectively solved. The experiment results demonstrated that compared with first come first serve（ FCFS）,particle swarm optimization（ PSO） and SA-PSO,LDD-SA-PSO algorithm has better ability of seeking optimization and convergence speed on the arrival flights scheduling,and the parameters of improved mathematical model also has obvious influence on the optimization results.

关 键 词：进港航班排序 最小延迟成本 数学模型 粒子群算法 线性微分递减 

分 类 号：V355[航空宇航科学与技术—人机与环境工程]