摘要
针对高超声速变形飞行器再入轨迹优化问题,研究了一种基于改进高斯伪谱法(Gauss pseudospectral method,GPM)的快速优化方法。首先,针对一种采用伸缩式机翼的高超声速变形飞行器,建立了将展长变形量扩展成为控制变量的再入轨迹优化模型。其次,采用GPM将轨迹优化问题转化为非线性规划(nonlinear programming,NLP)问题,并基于NLP偏导数的稀疏性推导目标函数梯度和约束Jacobian矩阵的高效计算方法。最后,优化求解了变形飞行器的最大横向航程、再入可达区、最大终端速度和最小飞行时间。仿真结果表明,推导的梯度计算方法可有效提高优化求解效率,变形飞行器相对于固定外形飞行器的性能更加优越,最大横向航程、可达区覆盖范围、最大终端速度和最小飞行时间等指标均有显著提升。
Aiming at the reentry trajectory optimization problem for hypersonic deformable vehicle,a fast optimization method based on improved Gauss pseudospectral method(GPM)is studied.Firstly,a reentry trajectory optimization model is established for a hypersonic deformable vehicle with retractable wing,which extends the span deformation as a control variable.Secondly,the GPM is used to transcribe the trajectory optimization problem into a nonlinear programming(NLP)problem,and based on the sparsity of the NLP partial derivative,the objective function gradient and the constrained Jacobian matrix are calculated efficiently.Finally,the maximum lateral range,reentry reachable area,maximum terminal speed and minimum flight time of the deformed aircraft are optimized.The simulation results show that the derived gradient calculation method can effectively improve the optimization solution efficiency.The performance of the deformed aircraft is better than that of the fixed-shape vehicle,and the maximum lateral range,reach area coverage,maximum terminal speed and minimum flight time have all been significantly improved.
作者
岳彩红
唐胜景
郭杰
王肖
张浩强
YUE Caihong;TANG Shengjing;GUO Jie;WANG Xiao;ZHANG Haoqiang(School of Aerospace Engineering, Beijing Institute of Technology,Beijing 100081, China;Tactical Weapons Division, China Academy of Launch Vehicle Technology, Beijing 100076, China;China North Industries Corporation, Beijing 100053, China)
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2021年第8期2232-2243,共12页
Systems Engineering and Electronics
基金
国家自然科学基金(11572036)资助课题。
关键词
高超声速飞行器
变形飞行器
轨迹优化
GAUSS伪谱法
稀疏性
hypersonic vehicle
deformable vehicle
trajectory optimization
Gauss pseudospectral method(GPM)
sparsity
