摘要
为解决基于航迹运行(Trajectory Based Operation,TBO)模式下大规模航迹战略冲突解脱问题,建立了综合考虑最小化冲突数量和航迹调整成本的双目标优化模型.设计了基于超体积指标的多目标合作协同进化算法(Indicator-based Cooperative Coevolution Multi-objective Evolutionary Algorithm,IBCCMOEA)进行模型求解,采用非支配排序遗传算法(Non-dominated Sorting Genetic Algorithm Ⅱ,NSGA-Ⅱ)进行子种群优化.同时,设计了自适应遗传算子加快算法收敛速度.分别采用中国航路网络繁忙时段442架和1014架航班历史飞行数据进行仿真验证,将所提算法与NSGA-Ⅱ,基于分解的多目标进化算法(Multi-objective Evolutionary Algorithm Based on Decomposition,MOEA/D)以及多目标合作协同进化算法(Cooperative Coevolution Multi-objective Evolutionary Algorithm,CCMOEA)进行对比分析.实验结果表明所提算法相比于其他算法收敛速度提升了7.54%,可以获得无冲突的飞行计划且航迹调整成本减少了约9.62%.所获得的结果可以满足TBO环境下的协同决策需求,算法收敛速度可以满足效率要求.
To solve the large-scale trajectory strategic conflict resolution problem under the conception of trajectory based operation(TBO),a bi-objective optimization model that aims to minimize both potential conflicts and trajectory adjustment cost was established.Meanwhile,an indicator-based cooperative co-evolution multi-objective evolutionary algorithm(IBCCMOEA)was implemented.Specifically,the representative solution is selected from the subpopulation according to the hyper volume(HV).The well-known non-dominated sorting genetic algorithm-Ⅱ(NSGA-Ⅱ)was employed to optimize each sub-problem,and the adaptive genetic operators were designed to speed up the convergence of the algorithm.The proposed algorithm was compared with NSGA-II,multi-objective evolutionary algorithm based on decomposition(MOEA/D)and cooperative co-evolution multi-objective evolutionary algorithm(CCMOEA)by using the real traffic data from Chinese air route network.The Pareto front of IBCCMOEA in two cases always dominates the results obtained by the other three algorithms.In small-scale scenario,IBCCMOEA improves the convergence speed by 7.54%compared to the others,and achieves a conflict-free flight plan with about 9.62%reduction in trajectory adjustment cost.In the large-scale scenario,the convergence speed of IBCCMOEA is also improved by 5.56%.The algorithm can meet the demand of collaborative decision making in TBO environment and can satisfy the efficiency requirements.
作者
徐满
胡明华
周逸
张颖
XU Man;HU Minghua;ZHOU Yi;ZHANG Ying(College of Civil Avaition,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)
出处
《哈尔滨商业大学学报（自然科学版）》
CAS
2023年第5期521-532,共12页
Journal of Harbin University of Commerce：Natural Sciences Edition
基金
国家自然科学基金项目资助(71971112)。
关键词
空中交通管理
四维航迹规划
冲突解脱
多目标优化
超体积
合作协同进化算法
air traffic management
4 dimensional trajectory planning
conflict avoidance
multi-objective optimization
hyper volume
cooperative co-evolution algorithm
