Analysis and Optimization of Multistage Tesla Valves by Computational Fluid Dynamics and a Multi‐Objective Genetic Algorithm

Abstract

AbstractThe multistage Tesla valve was optimized by minimizing the forward pressure drop (FPD) and maximizing the reverse pressure drop (RPD). First, computational fluid dynamics simulations were conducted; then, surrogate‐based optimization was done. Finally, an explicit correlation between the objective functions and the variables was developed to estimate the target functions directly. The results showed that decreasing the valve‐to‐valve distance over the hydraulic diameter (G/Dh) maximizes the diodicity (Di) and minimizes the FPD. When an increase in the RPD is desired, the Reynolds number (Re) should be increased, which leads to an increase in the FPD. The maximized Di value is 1.811 for Dh = 0.821, G/Dh = 0.993, N = 16, tan(α) = 0.848, and Re = 173.03. However, the minimum value of the FPD and the maximum RPD were found as 194.72 Pa and 352.69 Pa, respectively.