Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

Inverse Design of Two-Dimensional Airfoils Using Conditional Generative Models and Surrogate Log-Likelihoods

Journal Article · · Journal of Mechanical Design
DOI:https://doi.org/10.1115/1.4052846· OSTI ID:1980680
; ; ; ;

Abstract

This paper shows how to use conditional generative models in two-dimensional (2D) airfoil optimization to probabilistically predict good initialization points within the vicinity of the optima given the input boundary conditions, thus warm starting and accelerating further optimization. We accommodate the possibility of multiple optimal designs corresponding to the same input boundary condition and take this inversion ambiguity into account when designing our prediction framework. To this end, we first employ the conditional formulation of our previous work BézierGAN–Conditional BézierGAN (CBGAN)—as a baseline, then introduce its sibling conditional entropic BézierGAN (CEBGAN), which is based on optimal transport regularized with entropy. Compared with CBGAN, CEBGAN overcomes mode collapse plaguing conventional GANs, improves the average lift-drag (Cl/Cd) efficiency of airfoil predictions from 80.8% of the optimal value to 95.8%, and meanwhile accelerates the training process by 30.7%. Furthermore, we investigate the unique ability of CEBGAN to produce a log-likelihood lower bound that may help select generated samples of higher performance (e.g., aerodynamic performance). In addition, we provide insights into the performance differences between these two models with low-dimensional toy problems and visualizations. These results and the probabilistic formulation of this inverse problem justify the extension of our GAN-based inverse design paradigm to other inverse design problems or broader inverse problems.

Research Organization:
Univ. of Maryland, College Park, MD (United States)
Sponsoring Organization:
USDOE Advanced Research Projects Agency - Energy (ARPA-E)
DOE Contract Number:
AR0001216
OSTI ID:
1980680
Journal Information:
Journal of Mechanical Design, Vol. 144, Issue 2; ISSN 1050-0472
Publisher:
ASME
Country of Publication:
United States
Language:
English

References (24)

Inverse Problem Theory and Methods for Model Parameter Estimation book January 2005
Solving inverse problems using data-driven models journal May 2019
Multilayer feedforward networks are universal approximators journal January 1989
Least Squares Generative Adversarial Networks conference October 2017
Deep learning in nano-photonics: inverse design and beyond journal January 2021
Generative Model for the Inverse Design of Metasurfaces journal September 2018
Designing nanophotonic structures using conditional deep convolutional generative adversarial networks journal June 2019
Free-Form Diffractive Metagrating Design Based on Generative Adversarial Networks journal July 2019
Simulator-based training of generative neural networks for the inverse design of metasurfaces journal November 2019
Inverse design of two-dimensional graphene/h-BN hybrids by a regressional and conditional GAN journal November 2020
Inverse design of porous materials using artificial neural networks journal January 2020
Generative Adversarial Networks for Crystal Structure Prediction journal July 2020
Computational Creativity Via Assisted Variational Synthesis of Mechanisms Using Deep Generative Models journal September 2019
Inverse molecular design using machine learning: Generative models for matter engineering journal July 2018
Deep Learning Techniques for Inverse Problems in Imaging journal May 2020
Conditional generative adversarial network for gene expression inference journal September 2018
An Inverse Method for Optimizing Elastic Properties Considering Multiple Loading Conditions and Displacement Criteria journal September 2018
Globally Approximate Gaussian Processes for Big Data With Application to Data-Driven Metamaterials Design journal September 2019
A Case Study of Deep Reinforcement Learning for Engineering Design: Application to Microfluidic Devices for Flow Sculpting journal September 2019
Synthesizing Designs With Interpart Dependencies Using Hierarchical Generative Adversarial Networks journal September 2019
Deep Generative Design: Integration of Topology Optimization and Generative Models journal September 2019
3D Design Using Generative Adversarial Networks and Physics-Based Validation journal November 2019
Computational Optimal Transport: With Applications to Data Science journal January 2019
SU2: An Open-Source Suite for Multiphysics Simulation and Design journal March 2016

Similar Records

Related Subjects

Engineering