摘要
提出结合改进遗传算法和序列二次规划法的可重复使用运载器再入轨迹优化方法,发挥了遗传算法(GA)对初值不敏感和全局收敛性强以及序列二次规划(SQP)法收敛速度快和精度高等优点,弥补了遗传算法优化结果的随机抖动、序列二次规划法对初始值敏感、收敛半径小和容易陷入局部极值等不足.将改进的遗传模拟退火罚函数法用于全局搜索设计空间,序列二次规划法用于局部优化,直接配点法用于将最优控制问题离散为非线性规划问题.算例结果表明,在没有初始估计的情况下,能够得到高精度的全局最优解,证明了该算法的正确性和有效性,验证了该算法具有初值不敏感和鲁棒性好的优点.
An optimization method combining improved genetic algorithm with sequential quadratic pro- gramming was proposed for the design of reusable launch vehicle reentry trajectory. The advantages of be- ing insensitive to initial values and global convergence of genetic algorithm（GA）, and rapid convergence and high precision of sequential quadratic programming （SQP） were developed. The weakness including solution vibration of GA and small convergence radius, being sensitive to initial values and easy to fall into a local extremum of SQP was overcome. The improved genetic algorithm with simulated annealing penalty function was employed to globally search design space and sequential quadratic programming for local opti- mization, while the direct collocation method was used to discretize optimal control problem into nonlinear programming problem. A global high-precision solution can be obtained without initial guess. Results show the correctness, effectiveness, insensitive to initial values and good robustness of the algorithm.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2014年第1期161-167,共7页
Journal of Zhejiang University：Engineering Science
基金
国家"863"高技术研究发展计划资助项目(2008AAXXX103)
