Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
April 21, 2026Results in EngineeringOpen Access

Fast parametric three-dimensional physics-informed neural network approach for fluid flows in complex geometries

View Full Paper
Ask AI
Bookmark
Share

Authors

CTChermen TsgoevDSDanil SakharovMBMiron Bratenkov

Discussion

Loading...

Member takes

Overview

Fast parametric neural network predicts fluid flows in complex geometries, suggesting improved accuracy and efficiency.

Key Points

  • The aim is to enhance the predictive capacity of physics-informed neural networks for 3D fluid flows in complex geometries.
  • Modified physics-informed neural networks for 3D Navier–Stokes equations
  • Employing finite difference derivatives for training efficiency
  • Geometry-informed input features for better accuracy
  • Substantial reduction in relative approximation errors observed
  • Fast convergence achieved despite complex geometries
  • Accurate solutions across Reynolds number intervals from 1 to 200

Cite This Study

Tsgoev et al. (2026) studied this question.

synapsesocial.com/papers/69e713decb99343efc98d4c7https://doi.org/10.1016/j.rineng.2026.110582
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Three-Dimensional Physics-Informed Neural Network Simulation in Coronary Artery Trees2024 · 23 citations
  2. 2Physics-informed neural networks for solving incompressible Navier–Stokes equations in wind engineering2024 · 18 citations
  3. 3Critical Investigation of Failure Modes in Physics-informed Neural Networks2022 · 43 citations
  4. 4Extended Physics-Informed Neural Networks (XPINNs): A Generalized Space-Time Domain Decomposition Based Deep Learning Framework for Nonlinear Partial Differential Equations2020 · 1,153 citations