基于代理模型的升力式再入飞行器组网优化

Lifting Reentry Vehicle Constellation Optimization Based on Surrogate Model

研究升力式再入飞行器的组网优化问题。优化再入飞行器组网,使其在满足一定覆盖要求的同时,轨道高度、离轨速度冲量、飞行器总数等方面达到最优。针对再入飞行器对地覆盖区域计算量大且和时间变量相关的问题,提出用代理模型技术来近似计算覆盖区域,提高了计算效率;利用SOC星座设计方法对再入飞行器进行了组网设计,并用枚举分析的方法对组网星座进行了优化。最后用点覆盖仿真的方法对得到的最优解进行了验证,说明了上述设计方法的有效性,为飞行器再入组网优化提供了科学依据。

The paper focused on the lifting reentry vehicle constellation optimization, which aims at obtaining the best solutions with respect to coverage requirements and orbit altitudes, de - orbit impulses, the number of vehicles, etc. To deal with the large computation of the attainable covering domain and its dependence on time, a surrogate model was employed to approximate the domain efficiently. The reentry vehicle constellation was designed with the streets of coverage （SOC） approach, and optimizated using the enumeration analysis method. At last, the optimal solu- tions were verified utilizing the point coverage simulation method, and the simulation results validated the proposed method.