{"title":"Sensitivity analysis method for aeroelastic aircraft models","authors":"Anthony A. Giunta","doi":"10.1016/S1369-8869(99)00016-6","DOIUrl":null,"url":null,"abstract":"<div><p>A novel method has been developed for calculating gradients of aerodynamic force<span><span> and moment coefficients for an aeroelastic aircraft model. This method is intended for use in preliminary-level aircraft design which typically involves computationally expensive aerodynamic and structural analyses. This method uses the global sensitivity equations (GSE) to express the aero-structural coupling in an aircraft model. In addition, a reduced-order </span>modal analysis<span> approach is employed to condense the coupling bandwidth between the aerodynamic and structural models. Coarse-grained parallel computing is applied to reduce the wall-clock computational time of the expensive aerodynamic analysis<span> needed in this sensitivity analysis method. A supersonic transport aircraft model is examined in this study, subject to Mach 2.4 cruise flight conditions. Aerodynamic analysis is performed using a NASA-developed Euler/Navier-Stokes solver, and structural analysis is performed using commercial finite element analysis software. The GSE/modal analysis method is used to compute the sensitivity of the aerodynamic performance<span> of the aircraft subject to perturbations in the angle-of-attack, wing sweep angle, and wing thickness. Good agreement is obtained between gradients computed with the GSE/modal analysis approach and the same quantities computed using a traditional, computationally expensive, finite difference approach. A cost analysis demonstrates that the GSE/modal analysis method is more computationally efficient than the traditional approach if gradients are needed for two or more aircraft design parameters.</span></span></span></span></p></div>","PeriodicalId":100070,"journal":{"name":"Aircraft Design","volume":"2 4","pages":"Pages 207-230"},"PeriodicalIF":0.0000,"publicationDate":"1999-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/S1369-8869(99)00016-6","citationCount":"9","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Aircraft Design","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1369886999000166","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 9
Abstract
A novel method has been developed for calculating gradients of aerodynamic force and moment coefficients for an aeroelastic aircraft model. This method is intended for use in preliminary-level aircraft design which typically involves computationally expensive aerodynamic and structural analyses. This method uses the global sensitivity equations (GSE) to express the aero-structural coupling in an aircraft model. In addition, a reduced-order modal analysis approach is employed to condense the coupling bandwidth between the aerodynamic and structural models. Coarse-grained parallel computing is applied to reduce the wall-clock computational time of the expensive aerodynamic analysis needed in this sensitivity analysis method. A supersonic transport aircraft model is examined in this study, subject to Mach 2.4 cruise flight conditions. Aerodynamic analysis is performed using a NASA-developed Euler/Navier-Stokes solver, and structural analysis is performed using commercial finite element analysis software. The GSE/modal analysis method is used to compute the sensitivity of the aerodynamic performance of the aircraft subject to perturbations in the angle-of-attack, wing sweep angle, and wing thickness. Good agreement is obtained between gradients computed with the GSE/modal analysis approach and the same quantities computed using a traditional, computationally expensive, finite difference approach. A cost analysis demonstrates that the GSE/modal analysis method is more computationally efficient than the traditional approach if gradients are needed for two or more aircraft design parameters.