Scholarlink

<p>The rising interest in artificial neural networks (ANNs) stems from their exceptional ability to handle complex, highly nonlinear problems in fields like fluid dynamics and biotechnology. This study explores the peristaltic flow of a magnetohydrodynamic (MHD) Reiner–Philippoff fluid through a curved channel using an artificial intelligence (AI) approach, specifically the Levenberg–Marquardt method within a backpropagation neural network (LMM-BNN). The model incorporates the effects of Ohmic heating, a radial magnetic field, and viscous dissipation. The governing equations, simplified under long wavelength and low Reynolds number approximations, are solved numerically to generate a reference dataset. This dataset is then used to train, test, and validate the LMM-BNN model to predict the results of velocity profile, temperature distribution, skin friction, and heat transfer rate. The analysis is conducted for eight distinct scenarios involving variations in the Bingham number, Hartmann number, curvature parameter, Reiner–Philippoff fluid parameter, and Brinkman number. The LMM-BNN model demonstrates excellent predictive accuracy, with mean square error (MSE) values as low as <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(10^{ - 11}\)</EquationSource> <EquationSource Format="MATHML"><math> <msup> <mn>10</mn> <mrow> <mo>-</mo> <mn>11</mn> </mrow> </msup> </math></EquationSource> </InlineEquation> for some cases. Key findings indicate that the velocity profile is enhanced near the channel walls but suppressed at the center with increasing Hartmann and Bingham numbers. The fluid temperature rises with the Hartmann and Brinkman numbers but declines with increasing Bingham number. Furthermore, the heat transfer rate at the wall is augmented by the Hartmann number. The LMM-BNN scheme proves to be a highly efficient and accurate computational tool for analyzing this complex biomechanical transport problem.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Significance of thermal transport analysis through an artificial intelligence stochastic approach for MHD peristaltic propulsion of Reiner–Philippoff fluid

  • 期刊论文

  • 发表日期：

    2025年10月31日

查看全文
  • J. Iqbal,
  • F. M. Abbasi,
  • M. M. Alam

摘要

The rising interest in artificial neural networks (ANNs) stems from their exceptional ability to handle complex, highly nonlinear problems in fields like fluid dynamics and biotechnology. This study explores the peristaltic flow of a magnetohydrodynamic (MHD) Reiner–Philippoff fluid through a curved channel using an artificial intelligence (AI) approach, specifically the Levenberg–Marquardt method within a backpropagation neural network (LMM-BNN). The model incorporates the effects of Ohmic heating, a radial magnetic field, and viscous dissipation. The governing equations, simplified under long wavelength and low Reynolds number approximations, are solved numerically to generate a reference dataset. This dataset is then used to train, test, and validate the LMM-BNN model to predict the results of velocity profile, temperature distribution, skin friction, and heat transfer rate. The analysis is conducted for eight distinct scenarios involving variations in the Bingham number, Hartmann number, curvature parameter, Reiner–Philippoff fluid parameter, and Brinkman number. The LMM-BNN model demonstrates excellent predictive accuracy, with mean square error (MSE) values as low as \(10^{ - 11}\) 10 - 11 for some cases. Key findings indicate that the velocity profile is enhanced near the channel walls but suppressed at the center with increasing Hartmann and Bingham numbers. The fluid temperature rises with the Hartmann and Brinkman numbers but declines with increasing Bingham number. Furthermore, the heat transfer rate at the wall is augmented by the Hartmann number. The LMM-BNN scheme proves to be a highly efficient and accurate computational tool for analyzing this complex biomechanical transport problem.