基于遗传算法优化的风洞试验MDOE方法  

Research on MDOE Method of Wind Tunnel Test based on Genetic Algorithm Optimization

作　　者：江嘉伟 程起有[1] 樊枫[1] 陈卫星[1] 汪文涛 JIANG Jiawei;CHENG Qiyou;FAN Feng;CHEN Weixing;WANG Wentao(National Key Laboratory of Helicopter Aeromechanics,China Helicopter Research and Development Institute,Jingdezhen 333001,China)

机构地区：[1]中国直升机设计研究所直升机动力学重点实验室,江西景德镇333001

出　　处：《直升机技术》2024年第1期63-68,共6页Helicopter Technique

摘　　要：为提高风洞试验效率,现代试验设计(MDOE)方法正越来越频繁地用于各类飞行器风洞试验。然而,基于MDOE方法的风洞试验尽管理论上减少了试验点数量,有效缩短了风洞吹风时间,但也带来了试验数据精度下降的问题。为此发展了一种基于D-最优设计准则和二元三次多项式响应面模型的MDOE方法,同时引入遗传算法进一步提高响应面模型的拟合精度。结果表明,所开发的MDOE风洞试验方法仅需单变量(OFAT)方法30%左右的试验点数,吹风时间较OFAT方法大约缩短了60%;与最初拟合的多项式响应面模型相比,遗传算法优化后的响应面拟合精度提高了5%左右,内插得到的数据误差更小。In order to improve the efficiency of wind tunnel test,modern design of experiment(MDOE)method are being used more and more frequently in wind tunnel test of various types of aircraft.However,although the number of test points is reduced theoretically,the wind tunnel test based on MDOE method effectively shortens the wind tunnel blowing time,but also brings the problem of the accuracy of the test data.In this paper,a MDOE method based on D-optimal design criterion and binary polynomial response surface model was developed,and the genetic algorithm was introduced to improve the fitting precision of the response surface model.The results showed that the MDOE wind tunnel test method required only one factor at a time(OFAT)method of 30%,and the wind time was approximately reduced by 60%in the OFAT method,compared with the original polynomial response surface model,the data that was used in the response surface of the genetic algorithm was more closer to the data obtained by the OFAT method.

关 键 词：D-最优设计准则 多项式响应面模型 MDOE 风洞试验 

分 类 号：V211.74[航空宇航科学与技术—航空宇航推进理论与工程] V249