Skip to content
Open access

Cross-Geometry Flow-Field Prediction in Irregular Wellbores Using Parameterized Physics-Informed Neural Networks

Nov 2026 · Processes · Vol 14, pp. 3550 · 0 citations · 37 references

Abstract

Irregular borehole enlargement alters annular flow structures and can create low-velocity regions that impair hole cleaning. However, conventional computational fluid dynamics (CFD) requires repeated geometric reconstruction and numerical simulations for each wellbore configuration, while existing physics-informed models have limited capability for cross-geometry extrapolation. This study develops transfer-learning-based parameterized physics-informed neural networks (TLP-PINNs) for the rapid three-dimensional prediction of Herschel–Bulkley drilling-fluid flow in locally enlarged irregular wellbores. A Gaussian enlargement ratio is introduced as an explicit geometric input, while mass and momentum conservation equations, boundary conditions, adaptive loss weighting, extrapolation-distance-driven physics-point allocation, and partial-layer transfer learning are integrated into a unified framework. The source model generally maintains good in-range interpolation accuracy for the velocity field within the 10–50% enlargement range, although the pressure error shows greater case-to-case variability for unseen intermediate geometries. For the unseen 80% enlargement case, transfer learning reduces the relative RMSEs of axial velocity, velocity magnitude, and pressure from 14.0%, 13.5%, and 3.5% to 9.42%, 9.50%, and 1.34%, respectively, with corresponding Rp2 of 0.9668, 0.9663, and 0.9992. The transferred model also accurately captures the low-velocity zone within the enlarged section and the associated redistribution of the local pressure gradient, while reducing neural-network training time by 51.2% relative to full-domain retraining. These results demonstrate a practical cross-geometry framework for the rapid hydraulic assessment of irregular wellbores and provide detailed flow-field information for subsequent analyses of hole cleaning and particle transport.

Read PDF

Similar papers

Oct 2026

A deep Euler-moment model for gas–liquid two-phase pipe flow

Ultra-deep gas wellbores (length-to-diameter ratio > 60 000) lack a computationally efficient and physically consistent one-dimensional flow model that integrates both data-driven learning and conservation-law priors. This paper introduces a deep Euler-moment model leveraging a hybrid Physics-Informed Neural Network ar...

Xing-Yu Li, Hui Zhang, Kun-Hong Lv et al. · 0 citations
Oct 2026

A parametric physics-informed neural network framework for cylinder-induced thermal flows

Thermal design and monitoring of compact heat exchangers require fields across changing operating conditions, and increasingly from a few embedded sensors, yet each new Reynolds number is a new mesh-based simulation. A physics-informed neural network (PINN) framework is presented for the coupled prediction and inverse...

Gaston Haidak, Jian-Long Kou, Dong-Yun Wang · 0 citations
Nov 2026

Reduced-order physics-constrained deep learning for shock-dominated supersonic channel flows

A reduced-order physics-constrained deep learning framework, SupersonicNet, is developed for the rapid prediction of two-dimensional supersonic flow fields (Ma 2.5–5.0) in confined channels with variable forward-facing step geometries. The framework is built upon a modified U-Net architecture. The continuity equation a...

Yu-De Yan, Jian Dai · 0 citations
Jan 2027

Physics-Constrained Supersonic Flow Prediction across a Discrete Speed Ladder in a Confined Channel

Deep learning models show great promise for rapid computational fluid dynamics (CFD), yet the comparative effectiveness of data-driven and physics-informed approaches for complex high-speed flows has been unclear. We conducted a rigorous comparative analysis of purely data-driven and physics-constrained neural networ...

Yu-De Yan, Jian Dai · 0 citations
Preprint Oct 2026

Regional accuracy of physics-informed neural networks (PINNs) for recirculation in a deep lid-driven cavity

Deep lid-driven cavities contain strong motion beneath the lid and much weaker recirculation at depth, so agreement in bulk velocity is an incomplete assessment of a computed flow. We examine data-free physics-informed neural networks (PINNs) for steady, two-dimensional incompressible flow in a cavity of depth-to-width...

E. Roohi, Christopher J. McDevitt · 0 citations
#graph neural networks Open access Oct 2026

Learnable coarsening for two-phase flow in porous media

Hydrodynamic simulations of multi-phase flow in porous media are essential for oil production forecasting but require significant computational resources. Traditional approaches lead to high memory and time demands, restricting multi-scenario analysis. We propose a learnable coarsening algorithm based on graph neural n...

V. Naumov, N. Yavich, Alexander A. Ryabov et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.