Artificial intelligence framework for predicting inclined magneto-bioconvective Ellis penta-hybrid nanofluid flow over a stretching cylinder with gyrotactic microorganisms: A Levenberg–Marquardt approach

This study aims to develop data-driven framework for predicting the magneto-bioconvective transport characteristics of an Ellis penta-hybrid nanofluid over a stretching boundary, with particular emphasis on rheological behavior, Brownian diffusion, thermophoresis, and gyrotactic microorganism transport. The Levenberg–Marquardt backpropagation scheme was employed to optimize a two-layer feed-forward neural model comprising sigmoid-activated hidden units and a linear output layer. The resulting BLMS-ANN predictions showed close agreement with the numerical data, with MSE values ranging from 10 −9 to 10 −11 across the training, validation, and testing stages. Physically, increasing the Ellis fluid parameter enhanced the velocity profile while reducing both nanoparticle concentration and motile microorganism distributions. In contrast, increasing the thermophoresis parameter reduced the temperature profile but increased the concentration and microorganism profiles. These findings demonstrate the capability of the proposed ANN framework to reproduce strongly coupled transport behavior with low prediction error and reduced computational effort. The present results are relevant to engineering applications involving thermal management, biomedical transport, energy systems, and advanced heat-transfer devices employing non-Newtonian hybrid nanofluids.