航空学报 > 2016, Vol. 37 Issue (11): 3413-3424   doi: 10.7527/S1000-6893.2016.0053

基于搜索空间变换和序优化的预警星座设计

刘冰, 易泰河, 申镇, 易东云   

  1. 国防科学技术大学 理学院, 长沙 410072
  • 收稿日期:2015-10-23 修回日期:2016-02-29 出版日期:2016-11-15 发布日期:2016-03-02
  • 通讯作者: 易东云,男,博士,教授,博士生导师。主要研究方向:网络科学与大数据。Tel.:0731-84573206,E-mail:dongyunyi@sina.com E-mail:dongyunyi@sina.com
  • 作者简介:刘冰,男,博士研究生。主要研究方向:装备系统分析、评估与优化。Tel.:0731-84573260,E-mail:liubeing@126.com;易东云,男,博士,教授,博士生导师。主要研究方向:网络科学与大数据。Tel.:0731-84573206,E-mail:dongyunyi@sina.com
  • 基金资助:

    国家自然科学基金(61370013,91438202)

Missile warning constellation optimization based on search space transformation and ordinal optimization

LIU Bing, YI Taihe, SHEN Zhen, YI Dongyun   

  1. College of Science, National University of Defense Technology, Changsha 410072, China
  • Received:2015-10-23 Revised:2016-02-29 Online:2016-11-15 Published:2016-03-02
  • Supported by:

    National Natural Science Foundation of China (61370013, 91438202)

摘要:

高轨预警星座由若干颗地球静止轨道卫星和大椭圆轨道卫星组成,星座优化的目标是满足重点监视区域的覆盖和提高覆盖区域的定位精度。对于覆盖性优化,根据多个监视区域的威胁等级,星座系统需要提供不同的覆盖重数;对于定位精度优化,系统在立体观测和单星观测情况下存在很大差异。因此高轨预警星座优化是一个复杂多区域多目标优化问题。针对以上难点,提出了多层次多目标优化模型,可以较完整合理地描述预警星座优化问题;在优化模型求解方面,将优化计算分为覆盖性优化和定位精度优化两个环节;在覆盖性优化环节,提出了基于搜索空间变换的覆盖性快速优化方法,提高了Pareto最优解的计算速度和准确性。在定位精度优化环节,采用序优化方法进一步缩短优化时间。仿真试验表明,该方法可设计出满足预警任务需求的星座,且优化耗时在1 min以内,能有效地缩短预警星座优化时间。

关键词: 卫星, 星座, 区域覆盖, 多目标, 优化

Abstract:

High earth orbit missile warning satellite constellation is composed of several satellites operating on the geostationary orbit and highly elliptical orbit. The coverage requirement and positioning precision are the most important objectives of its constellation optimization. For the coverage optimization, the coverage requirement is determined by the threat degrees of different areas. For the positioning optimization, the positional precision in multi coverage and single coverage regions is different. These two facts define the constellation optimization as a complicated multi-regional multi-objective problem. To deal with this problem, a multi-layer multi-objective model which is able to describe the problem efficiently is proposed. Then a method which defines the optimization as a process with two steps is introduced to solve the problem. In the process of coverage optimization, a rapid method based on search space transformation is introduced to the calculation of Pareto optimal, which is time saving and improves the accuracy. In the process of positioning optimization, the ordinal optimization is proposed to save the time of optimization. Finally, the validity of the algorithm is tested by the numerical experiment. And the time consumption of the entire optimization is less than one minute.

Key words: satellites, constellation, regional coverage, multi-objective, optimization

中图分类号: