Abstract
Numerical and optimization study of jet impingement cooling of a partly curved surface with CuO-water nanofluid was performed with Galerkin weighted residual finite element method and COBYLA (constrained optimization by linear approximation) optimization algorithm. Target surface was partly curved which has a semi elliptic shape and kept at constant hot temperature. Simulations were performed for various values of Reynolds number and solid particle volume fraction. It was observed that effects of curved wall on the distribution of fluid flow and heat transfer characteristics are more pronounced for higher values of Reynolds number as compared to a flat wall configuration. Highest heat transfer is obtained with curved wall and significant differences are observed between the peak values of Nusselt number between a flat wall and curved wall case. The average Nusselt number is a linear increasing function of nanoparticle volume fraction and the trends in local and average heat transfer are similar for curved wall and flat wall configurations when nanoparticles are added. Average Nusselt number enhances by about 20% at the highest particle volume fraction as compared to water. A polynomial type correlation for the average Nusselt number was derived which depends on the Reynolds number and solid particle volume fraction for both configurations.