Nili Ahmadabadi M, Ghadak F, Mohammadi M Subsonic and transonic airfoil inverse design via Ball-Spine Algorithm. In: 41st Aerospace Sciences Meeting and Exhibit, American Institute of Aeronautics and Astronautics. Gardner B, Selig M Airfoil Design Using a Genetic Algorithm and an Inverse Method. M Tyan J-W Lee 2019 Efficient multi-response adaptive sampling algorithm for construction of variable-fidelity aerodynamic tables Chin J Aeronaut 32 3 547 558 TJ Mackman CB Allen 2010 Investigation of an adaptive sampling method for data interpolation using radial basis functions Int J Numer Methods Eng 83 7 915 938 Z-H Han S Görtz 2012 Hierarchical kriging model for variable-fidelity surrogate modeling AIAA J 50 9 1885 1896 Lee D, Nguyen NV, Tyan M, Chun HG, Kim S, Lee J-W (2016) Enhanced multi-fidelity model for flight simulation using global exploration and the Kriging method. M Tyan NV Nguyen J-W Lee 2015 Improving variable-fidelity modelling by exploring global design space and radial basis function networks for aerofoil design Eng Optim 47 7 885 908 V Sekar M Zhang C Shu BC Khoo 2019 Inverse design of airfoil using a deep convolutional neural network AIAA J 57 J Eason S Cremaschi 2014 Adaptive sequential sampling for surrogate model generation with artificial neural networks Comput Chem Eng 68 220 232 In: Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, Mar. Glorot X, Bengio Y Understanding the difficulty of training deep feedforward neural networks. L Wang Z Liu 2021 Data-driven product design evaluation method based on multi-stage artificial neural network Appl Soft Comput 103 107117 MA Ellafi LK Deeks RW Simmons 2021 Application of artificial neural networks to the design of subsurface drainage systems in Libyan agricultural projects J Hydrol Reg Stud 35 100832 ĪM Tawfik 2021 Design of channel section for minimum water loss using Lagrange optimization and artificial neural networks Ain Shams Eng J 12 1 415 422 Sharma, “Intelligent parameters optimization for laser cutting of highly reflective and thermally conductive materials using artificial neural network,” Mater. X Chen Z Yi Y Zhou P Guo SG Farkoush H Niroumandi 2021 Artificial neural network modeling and optimization of the solid oxide fuel cell parameters using grey wolf optimizer Energy 3459 KH Park SO Jun SM Baek MH Cho KJ Yee DH Lee 2013 Reduced-order model with an artificial neural network for aerostructural design optimization J Aircr 50 4 1106 1116 M Papadrakakis ND Lagaros Y Tsompanakis 1998 Structural optimization using evolution strategies and neural networks Comput Methods Appl Mech Eng 156 1 309 333 (97)00215-6 NV Nguyen J-W Lee M Tyan S Kim 2015 Repetitively enhanced neural networks method for complex engineering design optimisation problems Aeronaut J 119 1220 1253 1270 GG Wang S Shan 2006 Review of metamodeling techniques in support of engineering design optimization J Mech Des 129 4 370 380 Chichester, England Hoboken, N.J: Wiley, 2005.ĪIJ Forrester AJ Keane 2009 Recent advances in surrogate-based optimization Prog Aerosp Sci 45 1–3 50 79 įorrester A, Sobester A, Keane A (2008) Engineering design via surrogate modelling: a practical guide. Keane A, Nair P Computational approaches for aerospace design: the pursuit of excellence, 1 edition. The second one shows whether the IDNN can generate existing airfoils by a set of target parameters. The first case study shows how accurate the network can generate an airfoil with arbitrary input parameters. The accuracy of the Airfoil-IDNN was demonstrated using two case studies. During the database enhancement loop the validation mean squared error has reduced to 57.37%, and totally to 96.48% after the final hyperparameter selection. The Airfoil-IDNN generates a NACA 4-series airfoil that corresponds to target aerodynamic parameters given as an input. A construction of Airfoil inverse design neural network is demonstrated in this paper. After the training database is generated the final hyperparameter selection is performed. Then database enhancement loop refines the training database by iteratively adding new samples to initial database based on results of validation. The initial hyperparameter selection defines the configuration of IDNN that increases the accuracy of inverse design and reduces the network training time. The algorithm generates an inverse design deep neural network (IDNN) by performing a two-stage hyperparameter selection and iteratively enhancing training database. A deep-learning based rapid engineering design (DL-RED) algorithm is developed and presented for solving engineering inverse design problems.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |