Statistical neural network-based prediction of non-Fourier heat and mass transfer in Walter-B hybrid nanofluid flow over an inclined cylinder with convective boundary conditions
Chemical Papers, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Basım Tarihi: 2026
- Doi Numarası: 10.1007/s11696-026-05881-4
- Dergi Adı: Chemical Papers
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Chemical Abstracts Core, Materials Science & Engineering Collection (ProQuest), Technology Collection (ProQuest)
- Anahtar Kelimeler: Cattaneo-Christove flux model, Hybrid nanofluid, Inclined cylinder, Statistical neural network (SANN), Thermal radiation, Walter-B fluid
- Kocaeli Üniversitesi Adresli: Evet
Özet
The study deals with the steady boundary layer flow of non-Fourier Walter-B hybrid nanofluid over an inclined deformable cylinder when subjected to convective boundary conditions. The hybrid nanofluid is made up of copper (Cu) and silver (Ag) nanoparticles dispersed in a kerosene oil base fluid. The heat and mass flux models for non-Fourier Cattaneo-Christov are integrated, as are the thermal radiation, heat source or sink, and chemical reaction effects, so the entire model is included in the transport model. The velocity, temperature and concentration distributions are computed throughout the computational domain using a data-driven Statistical Neural Network (SANN) framework. The MATLAB bvp4c finite difference solver is used to produce high-precision reference solutions for network training, validation and testing. SANN’s performances are thoroughly assessed by extensive residual plots, error statistics, convergence features and linear regression coefficients. The predictive network shows the minimum validation mean squared error of 1.0484 × 10–7 at epoch 13 and the correlation is perfect (R = 1.0) in all datasets through regression analysis. Parametric evaluations show that the velocity profiles are improved by reducing the viscous resistance by using fluid and curvature parameters. The curvature increases the thermal transport, while the radiation and Biot number decrease the transport of heat through the wall, but both the thermal relaxation factor and the solutal one limit the respective transport region. It is found that the skin friction coefficient increases from 1.595874 to 5.038795 with an increase in the fluid parameter and the Nusselt number decreases from 1.684256 to 1.593492 with an increase in the thermal radiation parameter. Further, the mass transfer enhancement is obtained in the form of Sherwood number increases from 1.679796 to 1.721043 as the chemical reaction parameter increases, indicating better mass transfer at the solid surface boundary.