基于METRIC理论的超大规模卫星星座多级备份策略

针对超大规模卫星星座可能出现的运行维护成本高、任务可用度低的问题, 结合卫星批量化生产以及采用一枚火箭将多颗卫星送入轨道(一箭多星)的现状, 基于可维修备件多级保障库存技术(multi-echelon technique for recoverable item control, METRIC)提出一种超大规模卫星星座多级备份策略。首先, 阐述了基于METRIC理论的卫星星座多级备份策略模型的原理, 备份策略同时考虑地面备份、停泊轨道备份以及星座轨道备份3种备份模式, 建立了考虑卫星年平均需求量、期望需求量与期望短缺数的星座可用度模型, 包含制造费用和发射费用的星座备份费用模型，以及以星座可用度为约束, 星座备份费用为优化目标的数学优化模型。然后, 以某型超大规模低轨卫星星座为案例研究对象, 使用遗传算法和边际分析法进行求解。最后, 讨论了遗传算法与边际分析法在求解该优化问题时的优缺点。结果表明, 在同一星座可用度指标下, 采用星座多级备份策略可比采用传统两级备份策略有效降低每年备份费用; 在保证星座可用度的前提下大幅降低了超大规模卫星星座的备份费用, 可为精确配置各级备份卫星数量提供参考。

Super large-scale satellite constellation multi-level backup strategy based on METRIC theory

Aiming at the possible problems of high operation and maintenance cost and low mission availability of super large-scale satellite constellation, combined with the current situation of mass production of satellites and the use of one rocket to put multiple satellites into orbit (one rocket with multiple satellites), a multi-level backup strategy of super large-scale satellite constellation is proposed based on the optimization method of multi-echelon technology for recoverable item control (METRIC). Firstly, the principle of the multi-stage backup strategy model of a satellite constellation based on METRIC theory is explained. The backup strategy considers three backup modes: ground backup, parking orbit backup, and constellation orbit backup. The constellation availability model considers the satellites' average annual demand number, expected demand number, and expected shortage number. The constellation backup cost model includes manufacturing cost and launch cost, and the mathematical optimization model with constellation availability as the constraint and constellation annual backup cost as the optimization objective. Then, a case study of a super large-scale constellation is developed and solved by using both genetic algorithm and marginal analysis. Finally, the advantages and disadvantages of the genetic algorithm and marginal analysis in solving this optimization problem are discussed. Result shows that the model significantly reduces the backup cost of a large-scale satellite constellation while ensuring the constellation availability, and provides guidance for the accurate allocation of the number of backup satellites at each level.