New insights into the micromixer with Cantor fractal obstacles through genetic algorithm

Abstract

This work is mainly to combine fractal principle with multi-objective genetic algorithm, and the multi-objective optimization of the Cantor fractal baffle micromixer is carried out. At different Reynolds numbers (Res), the three-dimensional Navier–Stokes equation is employed to numerically analyze the fluid flow and mixing in the microchannel. We choose the ratio of the three parameters associated with the geometry of the micromixer as design variables, and take the mixing index and pressure drop at the outlet of the micromixer as two objective functions for optimization. For the parameter study of the design space, the Latin hypercube sampling (LHS) method is used as an experimental design technique, and it is used to select design points in the design space. We use the proxy modeling of the response surface analysis (RSA) to approximate the objective function. The genetic algorithm is used to get the Pareto optimal frontier of the micromixer. K-means clustering is used to classify the optimal solution set, and we select representative design variables from it. Through multi-objective optimization, when Re = 1 and 10, the optimized mixing efficiency of the micromixer increased by 20.59% and 14.07% compared with the reference design, respectively. And we also prove that this multi-objective optimization method is applicable to any Res.