摘要
针对变比冲发动机作用下燃料最优地球逃逸轨道设计问题,给出了一种基于间接法的组合设计算法。该组合算法包括初值猜测技术和边值问题求解器两个部分:利用最优解的必要条件在初始时刻小邻域内构造一组关于协状态初值的线性方程组,以此给出短周期任务的初始猜值,然后结合曲线拟合技术猜测长周期任务的初值;边值问题求解器通过构造一个与待求解两点边值问题相对应的无约束优化问题,引入Armijo收敛性条件对迭代步长进行线性回溯搜索校正,以保证算法的全局收敛性。针对不同任务周期最优逃逸问题进行数学仿真,仿真结果表明该组合算法收敛性好,迭代次数少,对小推力轨道设计具有良好的应用价值。
A combinatorial algorithm to design fuel-optimal earth escaping trajectories powered by low-thrust, variable specific impulse propulsion is presented based on the indirect method, which involves an initial guess method and a boundary value problem solver. In order to generate accurate estimates of the initial co-states, a system of co-states linear equations is established in the near neighborhood of initial time firstly, then curve fits are used to give initial guess for longer escaping mission. The boundary value problem adopts a modified Newton's method with Broyden's method for extrapolating the Jacobian array, besides a line search and backtracking method based on Armijo conditions is used to insure the global convergence of the whole algorithm. Demonstrated for short and long duration escaping trajectories respectively, the algorithm is validated to possess fast convergence, high efficiency and might be used widely in low-thrust trajectories optimization.
出处
《导弹与航天运载技术》
北大核心
2011年第5期24-28,37,共6页
Missiles and Space Vehicles
基金
国家自然科学基金(61005060)
微小型航天器技术国防重点学院实验室开放基金(HIT.KLOF.2009090)
关键词
变比冲发动机
小推力
燃料最优
两点边值
初值猜测
Variable specific impulse rocket
Low-thrust
Fuel-optimal
Two-point boundary value
Initial guess