A framework for process-level radiative design using magnetized bioconvection dynamics in ternary hybrid nanofluid: Intelligent Levenberg-Marquardt technique

Abstract

This work aims to investigate the effects of thermal radiation and LTNE on the chemical reactive flow of a ternary hybrid nanofluid over a sheet containing thermo-bioconvection and oxytactic microorganisms. The model, which use artificial neural networks (ANNs) to forecast and optimize viscosity, heat dissipation, and thermal conductivity, is ideal for sophisticated cooling systems, energy storage, and biomedical applications. The ANNs has been trained using the Levenberg-Marquardt technique. The effectiveness of the scheme is supported by a number of statistical measures, such as analysis of error histograms, regression index, and convergence analysis, which show a minimum level of the best performance value ( $6.01\times E^{-8}\mathrm{to}1.77\times E^{-7}$ ) for the comprehensive simulation of the proposed model. Its applications include nuclear reactors, solar energy harvesting, electronic cooling, and chemical processes that require precise heat regulation. The addition of oxytactic microorganisms enhances heat transfer dynamics, boosting system efficiency and sustainability. The numerical findings are shown as tables and graphs on a Bvp4c. The liquid phase thermal profile increases while the solid phase thermal profile decreases as the interphase heat transfer parameter values grow.