Machine Learning in Aerodynamic Shape Optimization

Abstract

Machine learning (ML) has been increasingly used to aid aerodynamic shape optimization (ASO), thanks to the availability of aerodynamic data and continued developments in deep learning. We review the applications of ML in ASO to date and provide a perspective on the state-of-the-art and future directions. We first introduce conventional ASO and current challenges. Next, we introduce ML fundamentals and detail ML algorithms that have been successful in ASO. Then, we review ML applications to ASO addressing three aspects: compact geometric design space, fast aerodynamic analysis, and efficient optimization architecture. In addition to providing a comprehensive summary of the research, we comment on the practicality and effectiveness of the developed methods. We show how cutting-edge ML approaches can benefit ASO and address challenging demands, such as interactive design optimization. Practical large-scale design optimizations remain a challenge because of the high cost of ML training. Further research on coupling ML model construction with prior experience and knowledge, such as physics-informed ML, is recommended to solve large-scale ASO problems.

arXiv comment103 pages, 47 figures, published by Progress in Aerospace Sciences

References in corpus (47)

• Scikit-learn: Machine Learning in Python
• Auto-Encoding Variational Bayes
• Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network
• Machine Learning for Fluid Mechanics
• Deep Learning for Anomaly Detection: A Review
• An Introduction to Variational Autoencoders
• A Tutorial on Principal Component Analysis
• Turbulence Modeling in the Age of Data
• An Introduction to Deep Reinforcement Learning
• A guide to convolution arithmetic for deep learning
• B-PINNs: Bayesian Physics-Informed Neural Networks for Forward and Inverse PDE Problems with Noisy Data
• Machine learning accelerated computational fluid dynamics
• Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations
• A Survey on Deep Semi-supervised Learning
• Long Short-Term Memory Based Recurrent Neural Network Architectures for Large Vocabulary Speech Recognition
• Informed Machine Learning -- A Taxonomy and Survey of Integrating Knowledge into Learning Systems
• Machine learning in cardiovascular flows modeling: Predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networks
• A Physics Informed Machine Learning Approach for Reconstructing Reynolds Stress Modeling Discrepancies Based on DNS Data
• Deep Learning Methods for Reynolds-Averaged Navier-Stokes Simulations of Airfoil Flows
• Artificial Neural Networks trained through Deep Reinforcement Learning discover control strategies for active flow control
• Prediction of Aerodynamic Flow Fields Using Convolutional Neural Networks
• Learning data driven discretizations for partial differential equations
• Scalable Training of Artificial Neural Networks with Adaptive Sparse Connectivity inspired by Network Science
• Active subspace methods in theory and practice: applications to kriging surfaces
• Flow over an espresso cup: Inferring 3D velocity and pressure fields from tomographic background oriented schlieren videos via physics-informed neural networks
• Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equations
• A Point-Cloud Deep Learning Framework for Prediction of Fluid Flow Fields on Irregular Geometries
• An Overview of Deep Semi-Supervised Learning
• Deep Neural Networks for Nonlinear Model Order Reduction of Unsteady Flows
• A Survey of Deep Reinforcement Learning in Video Games
• Kernel Principal Component Analysis and its Applications in Face Recognition and Active Shape Models
• Reinforcement-learning-based control of confined cylinder wakes with stability analyses
• Learning the aerodynamic design of supercritical airfoils through deep reinforcement learning
• Learned discretizations for passive scalar advection in a 2-D turbulent flow
• Surrogate Modeling of Aerodynamic Simulations for Multiple Operating Conditions Using Machine Learning
• Multi-condition multi-objective optimization using deep reinforcement learning
• Convergence of the EM Algorithm for Gaussian Mixtures with Unbalanced Mixing Coefficients
• Deep Reinforcement Learning for Intelligent Transportation Systems
• Unsupervised Learning of Anomaly Detection from Contaminated Image Data using Simultaneous Encoder Training
• A Lagrangian Dual-based Theory-guided Deep Neural Network
• Adaptive sampling of pressure distribution features for the airfoil statistical analysis
• Customized data-driven RANS closures for bi-fidelity LES-RANS optimization
• A practical tutorial on Variational Bayes
• Modeling and Optimization with Gaussian Processes in Reduced Eigenbases -- Extended Version
• Generating various airfoil shapes with required lift coefficient using conditional variational autoencoders
• Grassmannian Shape Representations for Aerodynamic Applications
• Asymmetric Impurity Functions, Class Weighting, and Optimal Splits for Binary Classification Trees

Cited by in corpus (14)

• Improving aircraft performance using machine learning: a review
• Study of transfer learning from 2D supercritical airfoils to 3D transonic swept wings
• A complete state-space solution model for inviscid flow around airfoils based on physics-informed neural networks
• Parametric Model Embedding
• Flowfield prediction of airfoil off-design conditions based on a modified variational autoencoder
• Multi-fidelity Bayesian Optimization: A Review
• Towards aerodynamic surrogate modeling based on $\beta$-variational autoencoders
• The Principle of Minimum Pressure Gradient: An Alternative Basis for Physics-Informed Learning of Incompressible Fluid Mechanics
• Fixed-energy inverse scattering with radial basis function neural networks and its application to neutron-alpha interactions
• Multi-Fidelity Machine Learning Applied to Steady Fluid Flows
• Mesh-Agnostic Decoders for Supercritical Airfoil Prediction and Inverse Design
• Parametric Shape Optimization of Flagellated Micro-Swimmers Using Bayesian Techniques
• Data-driven parameterization refinement for the structural optimization of cruise ship hulls
• Uncertainty-Aware Data-Based Method for Fast and Reliable Shape Optimization