EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Numerical and hybrid physics-informed neural network study of magnetohydrodynamic blood-based Jeffrey ternary hybrid nanofluid flow over a wedge

Krishnan D, Durgaprasad P

Frontiers in Artificial Intelligence · 17.11.2026 · 9

DOI: 10.3389/frai.2026.1960993 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Artificial Intelligence, online erschienen am 17.11.2026. Verfasst von Krishnan D und 1 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

The current study explores the magnetohydrodynamic flow, heat transfer, and entropy generation characteristics of a blood-based Jeffrey ternary hybrid nanofluid containing graphene oxide (GO), titanium dioxide (TiO 2 ), and iron oxide (Fe 3 O 4 ) nanoparticles over an impermeable wedge embedded in a porous medium. The mathematical model accounts for the effects of the magnetic field, porous-medium resistance, thermal radiation, heat generation/absorption, viscous dissipation, and Jeffrey fluid rheology.

Methods

Through appropriate similarity transformations, the governing partial differential equations are converted into a system of nonlinear ordinary differential equations. The transformed equations are solved numerically using a boundary value problem (BVP) solver and subsequently analyzed through a Hybrid Physics-Informed Neural Network (HPINN) framework. The proposed HPINN combines governing-equation residuals, boundary constraints, and sparse numerical anchor data to obtain physically consistent predictions. The effects of the magnetic parameter, porous-medium resistance parameter, radiation parameter, Brinkman number, heat source/sink parameter, and Jeffrey fluid parameter on the velocity, temperature, entropy generation, skin-friction coefficient, and Nusselt number are examined in detail.

Result

The results show that the momentum transport characteristics are strongly influenced by the magnetic, porous-medium resistance, and Jeffrey fluid parameters, whereas thermal radiation enhances thermal boundary-layer growth. Entropy generation increases with increasing Brinkman number and Jeffrey fluid parameter, indicating greater thermodynamic irreversibility, particularly near the wedge surface. In addition, the HPINN predictions show close agreement with the benchmark numerical solutions and demonstrate satisfactory predictive capability for the transport quantities studied.

Discussion

The proposed methodology presents an efficient and reliable computational tool for the analysis of non-Newtonian hybrid nanofluid transport phenomena in biomedical engineering, thermal management systems, and intelligent heat-transfer applications.

Zitieren

APA

D, K., & P, D. (2026). Numerical and hybrid physics-informed neural network study of magnetohydrodynamic blood-based Jeffrey ternary hybrid nanofluid flow over a wedge. Frontiers in Artificial Intelligence, 9. https://doi.org/10.3389/frai.2026.1960993

Vancouver

D K, P D. Numerical and hybrid physics-informed neural network study of magnetohydrodynamic blood-based Jeffrey ternary hybrid nanofluid flow over a wedge. Frontiers in Artificial Intelligence. 2026;9. doi:10.3389/frai.2026.1960993

BibTeX
@article{D2026,
  author = {D, Krishnan and P, Durgaprasad},
  title = {Numerical and hybrid physics-informed neural network study of magnetohydrodynamic blood-based Jeffrey ternary hybrid nanofluid flow over a wedge},
  journal = {Frontiers in Artificial Intelligence},
  year = {2026},
  volume = {9},
  doi = {10.3389/frai.2026.1960993}
}
Kostenlos starten →