Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 15, 2024

Credibility Assessment of Machine Learning-Based Surrogate Model Predictions on NACA 0012 Airfoil Flow

View Full Paper
Ask AI
Bookmark
Share

Authors

JKJared KirschWRWilliam J. RiderNFNima Fathi

Discussion

Loading...

Member takes

Overview

Computational evaluation demonstrates credibility assessment of deep neural network surrogates for NACA 0012 airfoil flow, highlighting best practices for aerodynamic modeling.

Key Points

  • Rigorous credibility evaluation confirms that surrogate models accurately predict aerodynamic coefficients for computational mechanics, reducing substantial simulation costs.
  • Computational fluid dynamics simulations train a deep neural network surrogate to estimate lift and drag across varying Reynolds numbers and angles of attack on a NACA 0012 airfoil.
  • Establishing systematic best practices supports the formal verification of deep neural network predictions, facilitating trustworthy surrogate model deployment in computational mechanics.

Cite This Study

Kirsch et al. (2024) studied this question.

synapsesocial.com/papers/68e6a273b6db643587625a2dhttps://doi.org/10.1115/vvuq2024-132964
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Application of SciML-Adapted PCMM to Deep Neural Network Surrogate Model Used for Aerodynamic Coefficient Prediction2025
  2. 2Critical Investigation of Surrogate Modeling Based on Simultaneous Physics-Informed Deep Learning of the High Reynolds Number Flow around Airfoils under Variable Angles of Attack2024
  3. 3A Dual-Step Deep Learning-Based Surrogate Model for Dynamic Stall Predictions2024
  4. 4Data-efficient AI models for aerodynamic coefficient prediction: a case study on NACA airfoils with minimal CFD runs2026 · 1 citations
  5. 5Nonlinear Surrogate Model Design for Aerodynamic Dataset Generation Based on Artificial Neural Networks2024 · 7 citations