A numerical study and statistical approach of the impact of nanofluids on mixed convection in a ventilated cavity

In contemporary power engineering and microelectronics, the efficiency of cooling systems is of crucial importance. To meet this requirement, specialized approaches and the use of nanofluids are employed to improve the heat dissipation of heat-generating components. This study presents a methodology based on a numerical simulation investigation and statistical analysis using the Response Surface Methodology (RSM) to estimate the average Nusselt number (Nu_avg) associated with mixed convection in a ventilated cavity. The study considered pure water and mixtures of nanoparticles (Cu, Ag, and TiO₂) as heat transfer fluids, exploring various values of the Richardson number (0.1 to 100) and volume fractions (0 to 8%). Analysis of Variance (ANOVA) and the quadratic mathematical model developed by RSM effectively predicted the results of the numerical simulation with a coefficient of determination R² close to 1. The results obtained from the 3D curves of the RSM show that the average Nusselt number (Nu_avg) increases significantly with the Richardson number. Conversely, an increase in the volumetric fraction leads to a slight decrease in Nu_avg. Furthermore, it is observed that the agent (Ag) generates higher Nu_avg values compared to other materials such as copper (Cu) and TiO₂. Combining the RSM method with the desirability function (DF) allows for achieving the optimal average Nusselt number (Nu_avg). Validation of the values proposed by the DF and those obtained by simulation shows a very small relative error, less than 0.13%.