1种基于Kriging近似模型的叶片罩量优化设计方法  

Optimization Design Method of Gravity Center Eccentricity for Blade Based on Kriging Surrogate Model

作　　者：刘海涛[1] 徐胜利[1] 王晓放[1] 宋洋[2] 

机构地区：[1]大连理工大学能源与动力学院,辽宁大连116024 [2]中航工业沈阳发动机设计研究所,沈阳110015

出　　处：《航空发动机》2015年第5期24-27,共4页Aeroengine

基　　金：燃气轮机工程重大项目资助

摘　　要：在航空发动机叶片设计过程中,需要进行叶片罩量优化来减小多种载荷引起的弯曲应力,改善其应力状况。为了提高叶片罩量优化设计效率,根据Kri gi ng近似模型和试验采样技术,提出了1种叶片罩量优化设计方法。利用序列采样方法逐步改善近似模型预测精度,然后在近似模型上进行全局寻优。结果表明:该方法简单易用,通过构造近似模型代替真实的物理模型,降低了计算成本,提高了优化效率。优化后的叶片最大等效应力减小了12.43%,有效地减小叶片的峰值应力。During aeroengine blade design process , the gravity center eccentricity of blade was optimized to decrease and improve bending stress caused by kinds of loadings. In order to improve the optimization design efficiency of the gravity center eccentricity of the blade, a kind of design method was proposed according to the Kriging surrogate model and experimental sampling technology. The prediction accuracy of the surrogate model was improved step by step using sequence sampling method in order to search global optima. Results show that the method is simple to reduce the cost calculation and improve the optimization efficiency by building the surrogate model instead of real physical model. The maximum equivalent stress of the optimized blade reduces 12.43%, which can effectively reduce the peak stress of the blade.

关 键 词：叶片 罩量 KRIGING 试验设计采样 优化设计 航空发动机 

分 类 号：V232.4[航空宇航科学与技术—航空宇航推进理论与工程]