临近空间固体动力飞行器发动机与轨迹一体化设计优化

Integrated Design Optimization for Motor and Trajectory of Near Space Solid Rocket Motor Powered Vehicle

关键词:  固体火箭发动机  飞行器  轨迹  一体化设计  优化设计  自适应算法

Key words:  Solid rocket motor  Vehicle  Trajectory  Integration design  Optimization design  Adaptive algorithm

摘要:

针对临近空间多级固体动力飞行器发动机与轨迹一体化设计优化问题，提出一种基于序列代理优化的高效设计方法。为了准确计算发动机的性能特性，对发动机进行了几何参数化建模，并针对复杂装药的燃面计算，提出了基于移动四面体的燃面计算算法。为了准确获得给定设计方案的最大航程，采用自适应Legendre-Gauss-Radau伪谱法进行计算。为了提高发动机与轨迹一体化设计优化效率，提出了基于Kriging代理模型的多采样点高效全局优化算法，并进行了数值验证。计算结果表明：该方法的优化结果满足全部设计约束，最大航程相比优化前提高了9.34%。该结果与传统参数优化算法的结果一致，但耗时目标和约束函数的计算次数显著减少，计算时间仅为遗传算法的4%。

Abstract:

An efficient method based on sequential surrogate-based optimization is proposed to realize the integrated design optimization for motor and trajectory of multi-stage solid rocket motor powered vehicle in near space. In order to calculate the performance of the motor accurately， the geometric parametric modeling of the motor is carried out， then a marching tetrahedron method was developed to compute burning surface regression for complex propellant grain. The adaptive Legendre-Gauss-Radau pseudospectral method is used to compute the maximum range of given design scheme accurately. To improve the optimization efficiency， a multi-sampling efficient global optimization method based on Kriging surrogate model is proposed and numerically verified. The results show that optimization solution solved by this method meets all the design constraints， and the maximum range increased by 9.34% compared with that before optimization. The solution is consistent with that of traditional parameter optimization algorithms， however， the calls number of time-consuming objective function and constraint functions are significantly reduced， and this method only takes 4% calculation time of the genetic algorithm.