Training a Neural-Network-Based Surrogate Model for Aerodynamic Optimisation Using a Gaussian Process

DOI10.1080/10618562.2022.2138863OpenAlexW4319981052MaRDI QIDQ5880409FDOQ5880409

Authors: Nahla Alhazmi, R. Tezaur, C. Farhat 

Publication date: 2 March 2023

Published in: International Journal of Computational Fluid Dynamics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1080/10618562.2022.2138863

Recommendations

• Multi-fidelity deep neural network surrogate model for aerodynamic shape optimization
• Numerical investigation of minimum drag profiles in laminar flow using deep learning surrogates
• A deep learning approach for efficiently and accurately evaluating the flow field of supercritical airfoils
• Surrogate convolutional neural network models for steady computational fluid dynamics simulations
• An application of neural networks to the prediction of aerodynamic coefficients of aerofoils and wings

zbMATH Keywords

supervised learningcross-validationGaussian processadaptive samplingactive learningpressure coefficientneural network regressioncambered NACA airfoil

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Basic methods in fluid mechanics (76M99) Flow control and optimization for compressible fluids and gas dynamics (76N25)

Cites Work

• Scikit-learn: machine learning in Python
• On the limited memory BFGS method for large scale optimization
• No unbiased estimator of the variance of K-fold cross-validation
• Random sampling with a reservoir
• The origins of kriging
• Bayesian learning for neural networks
• Near-optimal sensor placements in Gaussian processes: theory, efficient algorithms and empirical studies
• Title not available (Why is that?)
• Application of a three-field nonlinear fluid-structure formulation to the prediction of the aeroelastic parameters of an F-16 fighter
• Advanced Lectures on Machine Learning
• Universality of deep convolutional neural networks
• Many regression algorithms, one unified model: a review
• Applying Bayesian optimization with Gaussian process regression to computational fluid dynamics problems
• Wide neural networks of any depth evolve as linear models under gradient descent ^*

Cited In (1)

• Modeling design and control problems involving neural network surrogates

Uses Software

• L-BFGS
• SPACE
• Scikit
• PyTorch

This page was built for publication: Training a Neural-Network-Based Surrogate Model for Aerodynamic Optimisation Using a Gaussian Process

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5880409)