Deep Learning Model Improves Turbulence Simulations in Engineering Applications
Researchers have developed the Deep Algebraic Reynolds Stress Model (DARSM), a physics-informed deep learning closure for Reynolds-averaged Navier-Stokes (RANS) turbulence simulations that reduces average velocity prediction errors by 2–4 times over baseline RANS. The model embeds physical structure from Reynolds stress transport equations into a neural network, enabling end-to-end training through the governing equations via custom adjoint methods. It outperforms five established machine learning approaches and generalises across Reynolds numbers, unseen geometries, and different flow regimes without retraining.
Turbulent flow simulation is central to engineering and science but remains computationally prohibitive at high fidelity. RANS equations offer dramatic computational savings but require closure models to approximate unclosed stress terms, a long-standing challenge. DARSM addresses this by training a neural network to map flow invariants to parameters in an implicit algebraic Reynolds stress equation derived from first principles under the weak-equilibrium assumption, imposing physical constraints on the machine learning component. A key technical contribution is the derivation of adjoint equations that exploit the solver's implicit-explicit structure, circumventing failures of standard automatic differentiation on the stiff coupled system and enabling end-to-end optimisation that eliminates distribution shift. On benchmark cases—square-duct and periodic-hill flows—DARSM achieves 2–4× average error reduction over baseline RANS, with peak reductions of 12× on individual cases. Notably, a model trained only on attached, anisotropy-dominated square-duct flows successfully generalises to separated periodic-hill flows without retraining, representing a significant regime change. The work is a preprint posted to arXiv and has not yet undergone formal peer review.
What's missing
As a preprint, DARSM has not yet undergone peer review. The study benchmarks on two canonical flow geometries; performance on more complex three-dimensional industrial geometries (e.g., airfoils, turbomachinery) remains untested. The weak-equilibrium assumption underlying the algebraic stress formulation may limit applicability to strongly non-equilibrium flows.
What different sources said
- arXiv physicsCenter
Deep Learning-based Algebraic Reynolds Stress Closures for RANS Simulations of Turbulent Flows
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.