基于全局灵敏度方程的多层次弹道优化设计

Multilayer trajectory optimization design for surface-to-air missile based on global sensitivity equation method

为了使防空导弹优化的弹道更有效的拦截目标,依据多层次法理论,以杀伤概率最大为目标,设计导弹中制导段弹道和末制导段时间参数.整个模型分为3层,第1层是导弹中制导段弹道,第2层是导弹末制导段,第3层是战斗部毁伤目标.将一个层次视为一个子系统,多层次弹道优化问题实质上就是多学科设计优化问题.采用全局灵敏度方程方法计算参数之间的敏度,在系统层进行优化,最终可得到杀伤概率最大的弹道.最后进行了拦截空地导弹目标的计算示例,并采用了均匀实验设计方法和BP神经网络以减小计算量.结果证明全局灵敏度方程是求解多层次弹道优化设计的有效方法.

A muhilayer approach to the midcourse trajectory design and terminal phase time constants design for a surface-to-air missile to maximize the kill probability was presented. Three layers were discussed. The first layer represented the midcourse trajectory of the missile, the second layer represented the terminal homing phase, and the third layer was the warhead damaging process. A layer can be seen as a subsystem, the multilayer trajectory optimization problem is a essential multidisciplinary design optimization （MDO） problem. The global sensitivity equation （GSE） method was adopted to compute sensitivity derivatives with respect to design variables, which did optimization in system level and got the trajectory which maximizing the kill probability. At last a case of intercepting air-to-surface missile（ASM） was designed and optimized. Uniform design of experiments and BP neural network were used to avoid long time computing. The result is that GSE method is effetive for multilayer trajectory optimization design problem.