mlsanityData to 7 Oct 2026

Machine Learning in Aerodynamic Shape Optimization

arXiv 2202.07141 ↗First posted 15 Feb 2022cs.LG math.OC physics.flu-dyncited by 39

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 39 papers in the map (data through 7 Oct 2026). Builds on 61 of its 525 references that are in the map. First posted on arXiv 15 Feb 2022.

Builds on

61 of 525 references.

References · 61

  1. 2013Auto-Encoding Variational Bayescited by 3,980
  2. 2015Deep Reinforcement Learning with Double Q-learningcited by 1,704
  3. 2019An Introduction to Variational Autoencoderscited by 997
  4. 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 858
  5. 2021Physics-informed neural networks (PINNs) for fluid mechanics: A reviewcited by 470
  6. 2021Machine learning accelerated computational fluid dynamicscited by 461
  7. 2018Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Networkcited by 415
  8. 2019Machine Learning for Fluid Mechanicscited by 380
  9. 2017Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equationscited by 362
  10. 2016Deep Reinforcement Learning for Robotic Manipulation with Asynchronous Off-Policy Updatescited by 342
  11. 2020B-PINNs: Bayesian Physics-Informed Neural Networks for Forward and Inverse PDE Problems with Noisy Datacited by 329
  12. 2017Scalable Training of Artificial Neural Networks with Adaptive Sparse Connectivity inspired by Network Sciencecited by 325
38 more
  1. 2021A Survey on Deep Semi-supervised Learningcited by 250
  2. 2019Informed Machine Learning -- A Taxonomy and Survey of Integrating Knowledge into Learning Systemscited by 203
  3. 2018An Introduction to Deep Reinforcement Learningcited by 201
  4. 2016A guide to convolution arithmetic for deep learningcited by 182
  5. 2020Deep Learning for Anomaly Detection: A Reviewcited by 159
  6. 2017Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equationscited by 158
  7. 2018Deep Learning Methods for Reynolds-Averaged Navier-Stokes Simulations of Airfoil Flowscited by 147
  8. 2019Prediction of Aerodynamic Flow Fields Using Convolutional Neural Networkscited by 144
  9. 2018Independently Recurrent Neural Network (IndRNN): Building A Longer and Deeper RNNcited by 142
  10. 2018Survey of multifidelity methods in uncertainty propagation, inference, and optimizationcited by 141
  11. 2014A Tutorial on Principal Component Analysiscited by 139
  12. 2020An Overview of Deep Semi-Supervised Learningcited by 121
  13. 2021Pros and Cons of GAN Evaluation Measures: New Developmentscited by 121
  14. 2014Long Short-Term Memory Based Recurrent Neural Network Architectures for Large Vocabulary Speech Recognitioncited by 111
  15. 2019Deep Learning vs. Traditional Computer Visioncited by 94
  16. 2020Neural Mechanics: Symmetry and Broken Conservation Laws in Deep Learning Dynamicscited by 93
  17. 2019Machine learning in cardiovascular flows modeling: Predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networkscited by 76
  18. 2019A Survey of Deep Reinforcement Learning in Video Gamescited by 73
  19. 2019A review on Deep Reinforcement Learning for Fluid Mechanicscited by 65
  20. 2020A Probabilistic Graphical Model Foundation for Enabling Predictive Digital Twins at Scalecited by 58
  21. 2014Generalized Low Rank Modelscited by 55
  22. 2019Deep Gaussian Processes for Multi-fidelity Modelingcited by 48
  23. 2020A Point-Cloud Deep Learning Framework for Prediction of Fluid Flow Fields on Irregular Geometriescited by 44
  24. 2020Deep Neural Networks for Nonlinear Model Order Reduction of Unsteady Flowscited by 36
  25. 2018A Gentle Introduction to Deep Learning in Medical Image Processingcited by 34
  26. 2016Clustering Algorithms: A Comparative Approachcited by 28
  27. 2019Direct shape optimization through deep reinforcement learningcited by 28
  28. 2020Data-Driven Aerospace Engineering: Reframing the Industry with Machine Learningcited by 27
  29. 2017Application of Convolutional Neural Network to Predict Airfoil Lift Coefficientcited by 26
  30. 2018Deep Reinforcement Learning for General Video Game AIcited by 25
  31. 2015Polynomial-Chaos-based Krigingcited by 21
  32. 2017Gradient-enhanced kriging for high-dimensional problemscited by 15
  33. 2020Enhanced data efficiency using deep neural networks and Gaussian processes for aerodynamic design optimizationcited by 15
  34. 2018Restricted Boltzmann Machines: Introduction and Reviewcited by 14
  35. 2017Unsupervised Machine Learning for Networking: Techniques, Applications and Research Challengescited by 12
  36. 2020Learning the aerodynamic design of supercritical airfoils through deep reinforcement learningcited by 10
  37. 2021Deep Generative Model for Efficient 3D Airfoil Parameterization and Generationcited by 10
  38. 2021A practical tutorial on Variational Bayescited by 10

50 most cited of 61 All 61 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 5; 2024: 11; 2025: 12; 2026: 8.

39 papers.

Sort:

By area:

  1. 2022Improving aircraft performance using machine learning: a reviewcited by 19
  2. 2023Ship-D: Ship Hull Dataset for Design Optimization using Machine Learningcited by 12
  3. 2023Multi-fidelity Bayesian Optimization: A Reviewcited by 12
  4. 2023On the Generalization of PINNs outside the training domain and the Hyperparameters influencing itcited by 10
  5. 2022Parametric Generative Schemes with Geometric Constraints for Encoding and Synthesizing Airfoilscited by 6
  6. 2024Towards aerodynamic surrogate modeling based on β-variational autoencoderscited by 5
  7. 2024Implicitly Guided Design with PropEn: Match your Data to Follow the Gradientcited by 4
  8. 2025ONERA's CRM WBPN database for machine learning activities, related regression challenge and first resultscited by 4
  9. 2022A Manifold-based Airfoil Geometric-feature Extraction and Discrepant Data Fusion Learning Methodcited by 3
  10. 2024Enhancing Vehicle Aerodynamics with Deep Reinforcement Learning in Voxelised Modelscited by 3
  11. 2024A Survey on Design-space Dimensionality Reduction Methods for Shape Optimizationcited by 3
  12. 2024Generative Aerodynamic Design with Diffusion Probabilistic Modelscited by 3
27 more
  1. 2025Overtone: Cyclic Patch Modulation for Clean, Efficient, and Flexible Physics Emulatorscited by 3
  2. 2025MaNGO - Adaptable Graph Network Simulators via Meta-Learningcited by 3
  3. 20253DID: Direct 3D Inverse Design for Aerodynamics with Physics-Aware Optimizationcited by 3
  4. 2024Tailoring Generative Adversarial Networks for Smooth Airfoil Designcited by 2
  5. 2024Rapid aerodynamic prediction of swept wings via physics-embedded transfer learningcited by 2
  6. 2025Three-dimensional Deep Shape Optimization with a Limited Datasetcited by 2
  7. 2023Knowledge-embedded meta-learning model for lift coefficient prediction of airfoilscited by 1
  8. 2025Accelerated Airfoil Design Using Neural Network Approachescited by 1
  9. 2025Context-aware Learned Mesh-based Simulation via Trajectory-Level Meta-Learningcited by 1
  10. 2026Uncertainty-Aware Data-Based Method for Fast and Reliable Shape Optimizationcited by 1
  11. 2023Latent Task-Specific Graph Network Simulatorscited by 0
  12. 2024The Principle of Minimum Pressure Gradient: An Alternative Basis for Physics-Informed Learning of Incompressible Fluid Mechanicscited by 0
  13. 2024A Multi-Fidelity Methodology for Reduced Order Models with High-Dimensional Inputscited by 0
  14. 2024A mechanism-driven reinforcement learning framework for shape optimization of airfoilscited by 0
  15. 2024Jacobian-Enhanced Neural Networkscited by 0
  16. 2025FuncGenFoil: Airfoil Generation and Editing Model in Function Spacecited by 0
  17. 2025AutoTandemML: Active Learning Enhanced Tandem Neural Networks for Inverse Design Problemscited by 0
  18. 2025LFR-PINO: A Layered Fourier Reduced Physics-Informed Neural Operator for Parametric PDEscited by 0
  19. 2025Airfoil optimization using Design-by-Morphing with minimized design-space dimensionalitycited by 0
  20. 2025A Surrogate-Augmented Symbolic CFD-Driven Training Framework for Accelerating Multi-objective Physical Model Developmentcited by 0
  21. 2026Prediction of Steady-State Flow through Porous Media Using Machine Learning Modelscited by 0
  22. 2026Learning From Design Procedure To Generate CAD Programs for Data Augmentationcited by 0
  23. 2026Data-informed lifting line theorycited by 0
  24. 2026Towards a Foundation-Model Paradigm for Aerodynamic Prediction in Three-dimensional Designcited by 0
  25. 2026eXact-Prior Variational Autoencoder (X-VAE): Learning Data-Adaptive Gaussian Mixture Priors for Latent Distributionscited by 0
  26. 2026Knowledge-Constrained Shape Optimization with a Mixture-of-Experts Neural Operator for High-Confidence Designcited by 0
  27. 2026Cost-Accuracy Trade-offs: Neural Operator vs Classical Numerical Solvercited by 0