Neural network prediction of shear stress in two phase Casson nanofluid with Cattaneo Christov flux with dissipative heat and bioconvective motile microorganism

In augmenting the heat and solutal transport within advanced thermal system, the role of non-Newtonian nanofluids is vital and they shows substantial behavior owing to their outstanding electrical conduction attributes. The present work addresses the two-phase Casson fluid model by the incorporation of Cattaneo-Christov heat and solutal flux theory which eliminates the classical Fourier and Fick laws by accounting for relaxation time. Furthermore, the insertion of inertial drag, magnetic and Darcy dissipation along with bioconvection induced by motile microorganism enriches the flow phenomena. The utilization radiative heat and the chemical reaction are useful in several applications such as polymer extrusion, solar thermal collector, space technology, etc. The governing system is reformed into the dimensionless factors by employing suitable and proper transformation rules and thereafter, the mathematical model is handled by utilizing spectral quasilinearization method (SQLM). The proposed approach provides the faster convergence of spectral collocation with stability of quasi-linearization confirms the efficient handling of the problem. The role of diversified factors equipped with the model are presented graphically and elaborated briefly. Furthermore, the outcomes are employed to train a neural network system to predict shear rate at the wall with a large number of data i.e., for the simulation 10,000 data are used for the variation different factors.