Bayesian optimized neural network modeling for radiative flow of MHD hybrid nanoliquid with oxytactic microbes for biomedical applications

This work examines the impacts of Joule heating and viscous dissipation on the bioconvection flow of a CNTs-water hybrid nanofluid over a sheet while optimising for the Soret and Dufour effects. The complex relationships governing the system are simulated and comprehended using a neural network technique based on Bayesian Regularized Optimizers. In commercial and ecological uses, oxytactic and gyrotactic microbes are significant. While oxytactic microbes improve wastewater treatment by migrating into oxygen-rich areas and speeding up pollutant destruction, gyrotactic microorganisms aid mineral transport by lining up with fluid motion. While microbial activity improves mixing and transport, the introduction of CNTs-based hybrid nanofluids increases thermal conductivity, allowing sophisticated bioreactors and diagnostic systems. All things considered, the model enables bioengineering and biomedical applications that call for exact thermal and solutal control. To reduce error and get approximation results for various CNTs-water hybrid nanofluid's system scenarios, the generated synthetic data is subjected to the Bayesian-Regularized approach. The effectiveness of the Bayesian Regularized technique, developed for computational fluid dynamics of hybrid nanofluid, is demonstrated by learning curves on mean square error, regression analysis, transition state index, error distribution histogram, and fitness plots.