Multi objective optimization of cylindrical shape roughness parameters in a solar air heater

Abstract

Roughening the absorber plate of a solar air heater is a convenient technique to increase the rate of heat transfer. This study aimed to investigate the effect of discrete cylindrical shape roughness elements on the heat transfer and flow characteristics of a square duct solar air heater using Computational Fluid Dynamics (CFD) simulations. A multi objective optimization algorithm was applied to find the optimum values of four roughness parameters including the roughness diameter (d), height (e), longitudinal pitch (L), and transverse pitch (S). The objectives of the optimization were maximizing the Nusselt number (Nu) and minimizing the friction factor (f). The optimization procedure includes designing of experiments, generating response surfaces, and determining optimum roughness parameters. The optimal space filling design (OSFD) of experiment method was employed to investigate the effect of roughness parameters on the Nusselt number and friction factor with a minimum number of numerical experiments. Once the designed numerical experiments were performed, the Kriging response surface method was used to express the relationships between the Nusselt number and friction factor with the roughness parameters. Then, the non-dominated sorting genetic algorithm II (NSGA II) was applied to determine the optimum roughness parameters. The optimum values of roughness diameter, height, longitudinal pitch, and transverse pitch relative to the channel height (H) were found to be d/H = 0.055, e/H = 0.053, L/H = 0.192, and S/H = 0.158, respectively which yielded a thermo hydraulic performance parameter of 1.20.