Heat and mass transfer in blood-based ternary nanofluid flow through a porous tumour channel under local thermal non-equilibrium and chemical reaction: Physics-consistent machine-learning surrogates
Results in Engineering, cilt.32, 2026 (ESCI, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 32
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.rineng.2026.111930
- Dergi Adı: Results in Engineering
- Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus
- Anahtar Kelimeler: Arrhenius activation energy, Blood-based ternary nanofluid, Entropy generation, Machine learning surrogate, MHD porous channel flow
- İstanbul Gelişim Üniversitesi Adresli: Evet
Özet
This study presents a coupled numerical and machine-learning investigation of targeted drug delivery in a blood-based ternary hybrid nanofluid (Au, Fe3O4, SWCNT) flowing through a horizontal channel with symmetrically embedded porous tumour layers under a transverse magnetic field. The governing equations, which simultaneously couple magnetohydrodynamics, Local Thermal Non-Equilibrium (LTNE) solid–fluid energy exchange, Rosseland radiation, internal heat generation, and Arrhenius-type reaction kinetics, are solved by an in-house Fortran finite-volume solver. A parametric dataset of N=1499 simulations is generated, and the resulting surrogate family achieves strong held-out accuracy for heat- and mass-transfer targets, with the most challenging metric N s reaching R2=0.8938. Independent high-fidelity confirmation shows a maximum discrepancy of 0.042 in Sh‾porous and 5.2×10−4 in N s, enabling Pareto-type design screening at orders-of-magnitude lower cost than direct simulation. The Darcy number determines whether flow penetrates the tumour tissue or bypasses it through the open channel; the Hartmann number shifts the dominant entropy source from the tumour boundary toward Joule heating in the tumour interior; and the Lewis number dictates whether drug depletion spans the entire channel or remains confined to a thin wall layer. Arrhenius kinetics introduce strong thermal–concentration coupling: higher temperatures accelerate drug consumption while intensifying entropy generation, such that thermal and chemical performance cannot be optimised in isolation — a finding with direct implications for the rational design of magnetically and thermally actuated nanomedicine protocols. The main novelty of the present work is the integration of LTNE porous-tumour heat exchange, Arrhenius-type drug consumption, entropy-generation analysis, and physics-consistent ML-based Pareto screening within a single CFD-confirmed framework. Practically, this framework provides a rapid decision-support route for identifying operating conditions that enhance porous-layer drug uptake while controlling entropy-related thermodynamic losses.