Optimization of double-diffusive mixed convection for a hybrid nanofluid system: A numerical and statistical coupled approach

Abstract

There has been a lot of research on nanofluid convection, but not much on the impacts of double-diffusive mixed convection, hybrid nanofluids, and complex geometries like L -shaped cavities using combined statistical and numerical approaches. To fill that important gap, the aim is to investigate the effect of major nondimensional variables Ri , Br , and Le on the system’s performance in terms of heat and mass transfer. To achieve this, a statistical approach using RSM was employed, coupled with numerical computation relayed on the FEM for solving governing equations. The results for the average rate of heat transfer ( Nu avg ) and the average rate of mass transfer ( Sh avg ) exhibit a proportional correlation to Br and Ri , while an inversion was observed for Le . Moreover, the statistical study demonstrates that Nu grows with the increase in Ri and Br but drops for Le , whereas Sh rises with each of the three variables. The Nu avg increases by 9.05% as ϕ hnf is increased from 1% to 4%, but decreases by 17.01% as Le is increased from 0.01 to 10. The statistical assessment of the framework shows high R² values (98.75% for Nu avg and 96.57% for Sh avg ), ensuring the accuracy of the framework in predicting the output. This research gives helpful guidance for improving the transfer of heat and mass in nanofluid industries.