Abstract: Satellite data transmission and receiving is one of the two important stages on earth observation satellite executing the tasks of users, and it is an optimization problem with multi-time windows and multi-resources constrains. As the important component of data transmission system, the relay satellite can provide the possibility that the data can transmit all-weather and real-time. The paper mainly expounds these two jobs about the data relay data relay system. Firstly, we set a data transfer containing risk control scheduling model. Secondly, we combine the Tabu Search and Gene Algorithm to solve the model. Furthermore, we adopt distributed parallel computing strategy to improve the convergence rate and robustness of the algorithm. Finally, the distributed simulation system validates the effectiveness of the algorithm and data transfer scheduling model.

Key words: scheduling model of satellite data transfer, scheduling algorithm, risk control, distributed parallel computing

摘要： 作为对地观测卫星任务执行的两个重要阶段之一,数传接收的规划任务是一个具有多时间窗口、多优化目标和多资源约束的NP-Hard优化问题。中继星的引入为数据全天候近实时传输提供可能,同时也为数传规划提出新的问题。本文主要完成两项工作:第一,建立风险控制的卫星数传接收规划模型;第二,阐述基于遗传禁忌的模型求解方法,进一步采用分布式并行求解策略,改善了求解算法的收敛速度和鲁棒性。最后,通过STK提供基础仿真数据,验证了本文规划模型和求解算法的有效性。

关键词: 卫星数传规划模型, 规划求解算法, 风险控制方法, 分布式并行求解