<p>This work briefly inspects the impacts of activation energy and heat generation on the unsteady thermo-bioconvection flow of a trihybrid nanofluid around a spherical containing oxytactic bacteria. The LMB-ANN (Levenberg–Marquardt backpropagation artificial neural network) model offers a sophisticated computational intelligence method for examining the thermo-bioconvection flow of trihybrid nanofluids and tribological characteristics that contain oxytactic microorganisms and activation energy effects. Viscosity, thermal conductivity, and flow properties are all optimally predicted by this model, which effectively represents the nonlinear behavior of heat and mass transport. Its use is critical in the biomedical, industrial, and technical domains of cooling systems, lubrication technologies, and bio-inspired fluid mechanics, where accurate modeling of intricate singlephase interactions is necessary for energy economy and performance improvement. The Bvp4c method is used to solve the governing equations for the current model. The Levenberg–Marquardt approach has been employed to train the artificial neural network. Numerous statistical measures, including regression index, error histograms, analysis of correlation, and convergence analysis, support the scheme’s efficacy by demonstrating a minimum level of the best performance value (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">$4.5389\times {E}^{-8}$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mn>4.5389</mn> <mo>×</mo> <msup> <mrow> <mi>E</mi> </mrow> <mrow> <mo>-</mo> <mn>8</mn> </mrow> </msup> </mrow> </math></EquationSource> </InlineEquation> to <InlineEquation ID="IEq2"> <EquationSource Format="TEX">$1.7081\times {E}^{-7}$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mn>1.7081</mn> <mo>×</mo> <msup> <mrow> <mi>E</mi> </mrow> <mrow> <mo>-</mo> <mn>7</mn> </mrow> </msup> </mrow> </math></EquationSource> </InlineEquation>) for the thorough simulation of the suggested model.</p>

Computational intelligence of Levenberg–Marquardt backpropagation neural networks to study thermo-bioconvection flow and tribology of trihybrid nanofluid with oxytactic microorganisms

期刊论文
发表日期：

2025年9月11日

查看全文

Munawar Abbas,
Mohamed Medani,
Yahia Said,
Nashwan Adnan Othman,
Mustafa Bayram,
Barno Abdullaeva,
Abdullah A. Faqihi,
Nidhal Ben Khedher

摘要

This work briefly inspects the impacts of activation energy and heat generation on the unsteady thermo-bioconvection flow of a trihybrid nanofluid around a spherical containing oxytactic bacteria. The LMB-ANN (Levenberg–Marquardt backpropagation artificial neural network) model offers a sophisticated computational intelligence method for examining the thermo-bioconvection flow of trihybrid nanofluids and tribological characteristics that contain oxytactic microorganisms and activation energy effects. Viscosity, thermal conductivity, and flow properties are all optimally predicted by this model, which effectively represents the nonlinear behavior of heat and mass transport. Its use is critical in the biomedical, industrial, and technical domains of cooling systems, lubrication technologies, and bio-inspired fluid mechanics, where accurate modeling of intricate singlephase interactions is necessary for energy economy and performance improvement. The Bvp4c method is used to solve the governing equations for the current model. The Levenberg–Marquardt approach has been employed to train the artificial neural network. Numerous statistical measures, including regression index, error histograms, analysis of correlation, and convergence analysis, support the scheme’s efficacy by demonstrating a minimum level of the best performance value ( $4.5389\times {E}^{-8}$ $4.5389 \times E^{- 8}$ to $1.7081\times {E}^{-7}$ $1.7081 \times E^{- 7}$ ) for the thorough simulation of the suggested model.