A novel Bvp4c-based data-driven intelligent artificial neural network approach for analyzing gyrotactic microorganisms in triple-stratified Prandtl nanofluids

This paper highlights the Prandtl nanofluid flow past a linearly stretchable surface in the existence of nanoparticles and gyrotactic microorganisms. The Prandtl fluid is a non-Newtonian behaviour which is represented by an extra stress tensor. Similarity variables are used to transform the PDEs into a coupled system of non-linear ordinary differential equations. An artificial neural network (ANN) method is also employed to estimate and validate the solution profiles. The results show that the activation energy and chemical reaction greatly affect mass transport, while the heat source/sink strongly affects heat transfer. Moreover, the gyrotactic microorganisms affect the bioconvective response of the nanofluid and improve the stability of the system in given circumstances. The comparison of the numerical and ANN-reliant solutions shows high correspondence, which proves the reliability and efficiency of the suggested ANN framework. The findings may support the design engineering of bio-nanofluid thermal and bioconvective transport systems.