考虑发射失败的多中心多卫星发射任务协同规划方法

在中国未来卫星发射需求急剧增加和卫星发射中心发射能力有限的情况下,为多颗卫星协调发射中心和发射时间变得日趋困难。为解决大量卫星发射任务的协同规划问题,以发射成本最少、发射失败概率最低为优化目标,建立了多中心多卫星发射任务协同优化的多目标混合整数规划模型。基于非支配排序的多目标优化算法(non-dominated sorting genetic algorithm II, NSGA II)框架,设计了求解模型的多目标进化算法,提出了发射中心选择的整数编码方案,给出了基于启发式搜索的发射时间规划解码算法,并设计了染色体质量检查与修正算法。基于中国现有的4个卫星发射中心和可能面临的6类发射任务,设计了包含10颗卫星发射任务的小规模案例和30颗卫星发射任务的大规模案例,对模型和算法进行了仿真验证。实验结果表明该方法能有效解决多中心多发射任务协同规划问题。

Schedule of multiple satellite launch missions in multiple launch centers considering failures

It becomes more difficult to schedule launch centers and launch time for multiple satellites in China, with the increase of satellites launch demand and the limited capability of launch centers. A mixed-integer programming model of optimizing multiple launch missions in multiple launch centers is set up, aiming at the least cost and the lowest failure probability. Based on the non dominated sorting genetic algorithm II(NSGA-II) frame, a multiple objects evolution algorithm is designed and an integer coding method is put forward. A decoding algorithm for launch time scheduling based on the heuristic search is also proposed, and an algorithm to examine and revise the quality of chromosomes is designed. Finally, a small-scale case study with 10 missions and a large-scale case study with 30 missions are conducted to validate the model and algorithms, basing on the exiting 4 satellites launch centers and probable 6 types of launch missions. The results indicate the efficiency of the method.