遗传算法在固体发动机结构可靠性计算中的应用  

The application of an improved genetic algorithm in computing structural reliablity of SRM

作　　者：张庆雅[1] 汪亮[1] 孙东[1] 

机构地区：[1]西北工业大学航天工程学院,西安710072

出　　处：《固体火箭技术》2002年第2期13-17,共5页Journal of Solid Rocket Technology

摘　　要：通过对偶遗传算法及采用实数编码技术、交叉概率Pc 和变异概率Pm 的自适应调整 ,同时嵌套快速梯度运算 ,不仅加快了搜索 ,提高了收敛速度 ,且明显地改善了定位精度 ,有效地避免了求解有约束的非线性优化问题时采用惩罚函数法所存在的问题。用此方法计算了受极限状态方程约束的固体火箭发动机结构可靠性指标 β ,算例表明该算法对计算结构可靠性指标 β非常有效。An improved dual genetic algorithm is presented, which adopts real number coding and self-adaptive adjustment of across-over probability P c and mutation probability P m, and is embeded with fast grads algorithm.It not only increases convergence speed, but also improves the precision of position and resolves the problem resulted from solving nonlinear constrained optimization by punishment function method.The structural reliablity index β of a solid rocket motor restrained by the limit-state equation is calculated by using the improved genetic algorithm.The examples show that the algorithm is very effective.

关 键 词：遗传算法 结构可靠性 计算 固体推进剂 火箭发动机 

分 类 号：V435[航空宇航科学与技术—航空宇航推进理论与工程]