Özet
This paper proposes two novel multifidelity neural network architectures developed for high-dimensional inputs such as computational flowfields. We employed a two-dimensional flow-varying transonic supercritical airfoil problem while exploring and comparing our methods with the former “multifidelity deep neural networks” from the literature. We call these novel methods “modified multifidelity deep neural networks” and “multifidelity con-volutional neural networks.” The main objective of this study is to establish an advanced multifidelity prediction framework that can be applied to any computational data; however, here, we applied our methods to the prediction of aerodynamic coefficients using pressure coefficient fields around the airfoil. To generate the dataset, first, a coarse grid is employed using the SU2 Euler solver for low-fidelity data; then, a relatively finer grid is used to obtain the viscous solutions by using the Spalart–Allmaras turbulence model for the high-fidelity data. The performance metrics to compare the methods are determined as the test accuracy, the physical training time, and the size of the high-fidelity samples. The results demonstrate that the proposed multifidelity neural network architectures outperform the multifidelity deep neural networks in predictive modeling using high-dimensional inputs by improving the multifidelity prediction accuracy significantly.
Orijinal dil | İngilizce |
---|---|
Sayfa (başlangıç-bitiş) | 264-275 |
Sayfa sayısı | 12 |
Dergi | Journal of Aerospace Information Systems |
Hacim | 20 |
Basın numarası | 5 |
DOI'lar | |
Yayın durumu | Yayınlandı - May 2023 |
Bibliyografik not
Publisher Copyright:© 2023 by Melike Nikbay.
Finansman
of Multifidelity and Multidisciplinary Methodologies Integrating Sonic Boom, Aeroelasticity and Propulsion System for Supersonic Aircraft Design.” The second author would like to acknowledge the Istabul Technical University Scientific Research Projects Coordination Center (ITU-BAP) Scientific Research Projects Coordination Center for the support provided under the FHD-2023-44365 project titled “Multi-Fidelity and Multi-Disciplinary Design Optimization of a Low Boom Supersonic Transport Aircraft.” The authors would like to express their gratitude to General Electric Gas Power for the research grant titled “Development of Convolutional Neural Network Based Predictive Modeling Code.” The authors would like to acknowledge Scientific and Technological Research Council of Türkiye (TUBITAK) for the research grant provided under the 218M471 TUBITAK 1001 project titled “Development
Finansörler | Finansör numarası |
---|---|
General Electric Gas Power for the research | |
Türkiye Bilimsel ve Teknolojik Araştırma Kurumu | 218M471 TUBITAK 1001 |
Istanbul Teknik Üniversitesi | |
Bilimsel Araştırma Projeleri Birimi, İstanbul Teknik Üniversitesi | FHD-2023-44365 |