Enhancing Computational Fluid Dynamics with Machine Learning
Research abstract by Ricardo Vinuesa (@Ricardo Vinuesa) from KTH!! Twitter: @ricardovinuesa
In this video we discuss the recent article published in Nature Computational Science by Ricardo Vinuesa and Steve Brunton, where the potential of machine learning (ML) to improve numerical simulations of fluid flows is discussed. In particular, we show the possibilities enabled by ML in the context of 1) accelerating direct numerical simulations (DNSs); 2) improving turbulence modeling for large-eddy simulations (LESs) and Reynolds-averaged Navier–Stokes (RANS) simulations; and also enhancing the development of reduced-order models (ROMs). Focusing on the latter, we describe a deep-learning framework for learning a minimal and near-orthogonal set of non-linear modes in the context of turbulent flows. In particular, we focus on a high-fidelity numerical database of a simplified urban environment. The proposed technique relies on β-variational autoencoders (β-VAEs) and convolutional neural networks (CNNs), which enable extracting non-linear modes while encouraging the learning of statistically-independent latent variables and penalizing the size of the latent vector. We demonstrate that by constraining the shape of the latent space, it is possible to motivate orthogonality and extract a set of parsimonious modes which enable high-quality reconstruction. This method exhibits an excellent performance in the reconstruction against linear-theory-based decompositions, where the energy percentage captured by the proposed method from five modes is equal to % against % from proper-orthogonal decomposition (POD).
Link to article on ML and CFD:
Link to article on non-linear orthogonal modal decomposition:
3 views
5
0
2 months ago 00:00:00 1
. Lost Alpha Enhanced Edition
2 months ago 00:00:35 1
Prince of Persia The Lost Crown - PC Distribution Update Trailer
2 months ago 00:16:59 1
Nobody Wants to Die | PS5 - Xbox Series S/X - PC | Graphics Comparison | Analista De Bits
2 months ago 01:26:01 1
Music for Work — Deep Future Garage Mix for Concentration
2 months ago 00:02:35 1
CEX vs DEX 🔥 Which is Better? | BLUM is a Hybrid Crypto Exchange | Blum Academy
2 months ago 00:00:35 3
Hyperkin DuchesS, an Officially Licensed S Controller Replica for Xbox Series X | S | One and PC
2 months ago 00:18:36 1
The U.S. Is Terrified of China’s Hypersonic Tech
2 months ago 00:00:46 1
KOF’98: Intro: Games
3 months ago 00:16:09 1
Mortal Kombat Komplete PC Mileena Ladder Playthrough
3 months ago 00:16:29 1
HOW TO IMPROVE GRAPHICS IN GTA 5 _ HOW TO INSTALL NVE QUANTV 3.0 LYVELY WORLD EXPANSION
3 months ago 00:04:17 1
IMPROVED GRAPHICS IN GTA 5 USING NVE + QUANTV 3.0 + LYVELY WORLD EXPANSION
3 months ago 00:17:43 1
Fedora 41 drops X11, GNOME loses director, Copilot lawsuit dismissed: Linux & Open Source News
3 months ago 02:40:02 1
The First Descendant - All Bosses (With Cutscenes) 4K 60FPS UHD PC
3 months ago 00:33:22 1
The Newest Variant of the Navy SEALs Lightweight Assault Machine Gun
3 months ago 00:28:31 1
Project Golden Dragon 2/3
3 months ago 00:31:10 1
Project Golden Dragon 1/3
3 months ago 00:32:12 1
Project Golden dragon 3/3
3 months ago 00:11:11 1
How principled coders outperform the competition
3 months ago 00:09:16 1
Russia’s Cutting-Edge Chip Design & Material Science Mastery
3 months ago 00:02:48 1
World of Warcraft Goes Unreal | Turtle WoW 2.0 Unreal Engine Trailer Classic+
3 months ago 00:04:14 1
Most Popular Operating Systems (Desktop & Laptops) 1978 - 2023
3 months ago 00:49:04 1
Scientists Discuss the Future of Biological Computing