Artificial neural networks for mass transfer and bioconvection analysis in radiative Eyring-Powell flow over a convective cylinder surface: Application to microbial fuel cells

Abstract

Aim and significance of the present study: This study explores bioconvection in fluid flows in circular cylinders; it adds new features and broadens the scope of applications, such as microbial interactions in cylinder flows. Assumptions considered in the present study: The governing nonlinear partial differential equations were framed considering thermal radiative, Eyring Powell, mixed convection, chemical processes, viscous dissipation, and motile microbes, along with no-slip, convective boundary conditions, and these governing equations were transformed into dimensionless forms using similarity transformations. Methodology: The partial differential equations solver function in the Maple version 2024 program is used to derive numerical solutions, and artificial neural networks are used to forecast true values using the matrix laboratory software. Conclusion: This research explores several parameter features to illustrate potential technological uses. Artificial Neural Networks could predict the truth values accurately, with coefficient of determination almost approaching 1, representing the prediction accuracy and results shown in Table-1 were obtained using the Levenberg-Marquardt algorithm with backward propagation through matrix laboratory; for the majority of findings, the discrepancy between the truth values and anticipated values is less than 5 %, suggesting that both approaching very close. This demonstrates how accurately and closely the Artificial Neural Networks forecasts match the numerical results; Biotechnology, electrical engineering, cancer treatment, missile technology, and biomedicine are all significantly impacted by this research.