基于分解策略的SSO发射轨道遗传全局优化设计

提出了基于轨道分解优化和遗传算法（GA）的SSO发射轨道优化设计策略。针对多个轨道段相互耦合问题,基于分解优化策略,将整个发射轨道设计问题分解为两个轨道段设计问题。为了高效可靠地获得全局最优解,对基本遗传算法进行了改进。首先提出了基于多变异操作等改进措施的改进遗传算法;此外,结合遗传算法的全局搜索特性和Powell算法的局部搜索特性,设计了一种串行混合遗传算法。一个二级SSO运载火箭的计算结果表明,轨道分解优化策略确保了问题的成功求解,改进遗传算法和混合遗传算法均可稳定地获得全局最优解,但是混合算法更有效地提高了GA性能。

Decomposition Approach and Genetic Algorithm Based Global Optimization of Launch Trajectory for Sun Synchronous Orbit

A methodology for Sun Synchronous Orbit (SSO) launch trajectory optimization design is proposed, which is based on trajectory decomposition optimization and Genetic Algorithm(GA).In order to solve the problems caused by coupling of multiple segment trajectories, the all-up trajectory optimization design problem is divided into two segment trajectory optimization design problems based on the decomposition optimization approach. In order to obtain the global solution efficiently and robustly, the basic GA is improved. First, an improved GA including some improvement measures such as multiple mutation operators is proposed.Furthermore, a pipeline hybrid GA combining the global search properties of GA with the local search characteristics of Powell algorithm is developed. The computational results of a two-stage SSO launch vehicle show that the trajectory decomposition optimization approach guarantees the problem to be solved successfully, and that the improved GA and the hybrid GA both locate the global solution reliably. However, it should be pointed out that hybridizing algorithm improves the performance of GA more effectively.