Enhanced heat and mass transfer are more important in biomedical engineering and industrial applications. The incorporation of multiple nanoparticles into base fluids is studied because their superior thermophysical properties enhance heat and mass transfer. Therefore, this study investigates heat and mass transfer during electromagnetohydrodynamic (EMHD) peristaltic transport of Casson blood-based tetra-hybrid nanofluid (TTHNF) flow through a curved channel, with particular attention to electroosmotic wall effects. The model is relevant to biomedical pumping systems, blood flow in arteries, targeted drug delivery, and microfluidic devices. The main objective of this study is to examine the blood flow behavior and thermodynamic characteristics of Casson TTHNFs under electroosmotic and MHD effects in a curved channel. In addition, artificial neural networks (ANNs) and response surface methodology (RSM) are employed as supporting tools for fast prediction, optimization, and sensitivity evaluation of heat and mass transfer rates. The novelty of the present study is to investigate the TTHNFs made of uranium dioxide (UO2), gold (Au), molybdenum disulfide (MoS2), and iron oxide (Fe3O4) dispersed in blood. The analysis is performed employing the Casson fluid model, which considers the Hall current, Joule heating, electroosmotic, porous medium, activation energy, and thermal radiation effects. The governing equations are transformed into nonlinear ordinary differential equations (ODEs) with the help of the lubrication theory and Debye–Huckel approximation. The simplified system is solved numerically by employing a shooting-based technique through Mathematica. To analyze the heat transfer rates (HTRs) and mass transfer rates (MTRs) for the peristaltic motion of TTHNFs through a curved channel, a new statistical approach, RSM has been used. ANN models are also used in investigating the dynamics of flow. Sensitivity analysis is also carried out to examine the HTRs and MTRs of TTHNFs. Furthermore, TTHNFs give the most significant influence in HTRs and MTRs for the base fluid, nanofluids (NFs), hybrid nanofluids (HNFs), ternary hybrid nanofluids (THNFs), and tetra-hybrid nanofluids (TTHNFs). One can notice that RSM and ANN models are highly reliable for this study, as R2, R2 (adj), and R2 (pred) resulted in exceptionally high values of HTRs and MTRs of TTHNFs. Specifically, the developed RSM model resulted in R2, R2 (adj), and R2 (pred) as 96.00%, 97.00%, and 96.99%, respectively, for HTRs and 96.98%, 95.97%, and 95.90%, respectively, for the MTRs. Additionally, the results of the present examination indicate that the velocity profile enhanced near the lower channel walls with an increase in the Debye–Huckle and curvature parameters. The tabular results indicate that HTRs increases with elevated values of the Joule heating parameter, Hartmann number, thermophoresis, and curvature parameter, whereas a decreasing trend is noted for the Casson and radiation parameters.

ANN-assisted RSM modeling and sensitivity analysis for thermodynamic features of electrokinetic peristalsis blood flow with multifunctional nanoparticles

期刊论文
发表日期：

2026年6月17日

查看全文

J. Iqbal,
Y. Akbar

摘要

Enhanced heat and mass transfer are more important in biomedical engineering and industrial applications. The incorporation of multiple nanoparticles into base fluids is studied because their superior thermophysical properties enhance heat and mass transfer. Therefore, this study investigates heat and mass transfer during electromagnetohydrodynamic (EMHD) peristaltic transport of Casson blood-based tetra-hybrid nanofluid (TTHNF) flow through a curved channel, with particular attention to electroosmotic wall effects. The model is relevant to biomedical pumping systems, blood flow in arteries, targeted drug delivery, and microfluidic devices. The main objective of this study is to examine the blood flow behavior and thermodynamic characteristics of Casson TTHNFs under electroosmotic and MHD effects in a curved channel. In addition, artificial neural networks (ANNs) and response surface methodology (RSM) are employed as supporting tools for fast prediction, optimization, and sensitivity evaluation of heat and mass transfer rates. The novelty of the present study is to investigate the TTHNFs made of uranium dioxide (UO₂), gold (Au), molybdenum disulfide (MoS₂), and iron oxide (Fe₃O₄) dispersed in blood. The analysis is performed employing the Casson fluid model, which considers the Hall current, Joule heating, electroosmotic, porous medium, activation energy, and thermal radiation effects. The governing equations are transformed into nonlinear ordinary differential equations (ODEs) with the help of the lubrication theory and Debye–Huckel approximation. The simplified system is solved numerically by employing a shooting-based technique through Mathematica. To analyze the heat transfer rates (HTRs) and mass transfer rates (MTRs) for the peristaltic motion of TTHNFs through a curved channel, a new statistical approach, RSM has been used. ANN models are also used in investigating the dynamics of flow. Sensitivity analysis is also carried out to examine the HTRs and MTRs of TTHNFs. Furthermore, TTHNFs give the most significant influence in HTRs and MTRs for the base fluid, nanofluids (NFs), hybrid nanofluids (HNFs), ternary hybrid nanofluids (THNFs), and tetra-hybrid nanofluids (TTHNFs). One can notice that RSM and ANN models are highly reliable for this study, as R², R² (adj), and R² (pred) resulted in exceptionally high values of HTRs and MTRs of TTHNFs. Specifically, the developed RSM model resulted in R², R² (adj), and R² (pred) as 96.00%, 97.00%, and 96.99%, respectively, for HTRs and 96.98%, 95.97%, and 95.90%, respectively, for the MTRs. Additionally, the results of the present examination indicate that the velocity profile enhanced near the lower channel walls with an increase in the Debye–Huckle and curvature parameters. The tabular results indicate that HTRs increases with elevated values of the Joule heating parameter, Hartmann number, thermophoresis, and curvature parameter, whereas a decreasing trend is noted for the Casson and radiation parameters.