Thermal analysis of MHD non-Newtonian nanofluid flow across a Riga parallel plates with CattaneoChristov heat flux: A deep learning approach

This study investigates the time-dependent heat and mass transfer in magnetohydrodynamic (MHD) non-Newtonian nanofluid flow between Riga parallel plates. In order to get precise predictions of transient heat conduction and to enhance the stability of nanofluids, the CattaneoChristov heat flux model and thermophoretic particle deposition are incorporated. Partial differential equations governing the system are transformed into ordinary differential equations and solved via the 4th-5th order Runge-Kutta-Fehlberg method, complemented by a deep learning-based analysis of engineering factors under various inputs. A comprehensive analysis of velocity, temperature, concentration, Nusselt number, skin friction, and Sherwood number under various parameters is conducted. Results reveal that the squeezing constraint reduces thermal and mass profiles, while the modified Hartmann number improves fluid behavior near the lower plate but diminishes it at the upper plate. Heat source/sink parameters enhance thermal profiles, while thermal relaxation and thermophoretic constraints reduce them. The rate of heat transfer enhances approximately about 32 % from viscous to nanofluids. The findings highlight the model's accuracy in predicting temperature and concentration profiles, offering valuable insights for advancing heat transfer efficiency in solar energy systems, with broad implications for thermal engineering and nanotechnology.