Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
MULTI-OBJECTIVE SHAPE OPTIMIZATION OF A PASSIVE MICROMIXER USING GENETIC ALGORITHM AND MULTI-OBJECTIVE DECISION-MAKING ALGORITHMS
Submission Author:
Eduardo Henrique Taube Cunegatto , RS
Co-Authors:
Eduardo Henrique Taube Cunegatto, Kristian Nascimento Telöken, Lenon Audibert Cisco, Flavia Zinani, Sandro José Rigo
Presenter: Eduardo Henrique Taube Cunegatto
doi://10.26678/ABCM.COBEM2023.COB2023-1945
Abstract
In this work, the Constructal Design Method (CDM), based on the Constructal Law, combined NSGA-II algorithm, is used to optimize the shape of a micromixer, with five degrees of freedom, in terms of mixing percentage and pressure drop for a Peclet number of 500. The systems are modeled by Computational Fluid Dynamics (CFD) solutions, using Navier-Stokes and species equations to predict the flow and mixing of water-dye mixture. The simulations were designed via Central Composite design method, and the results were used to create second-order metamodels for mixing percentage and pressure drop, employed in the multi-objective optimization via NSGA-II algorithm. With the best set of solutions (Pareto curve) defined, the multi-criteria decision-making algorithms (LINMAP, TOPSIS, and VIKOR) were employed. The results showed that the shape with the highest mixing percentage also has the highest pressure drop. Using the decision methods to choose an optimized solution, it was possible to find shapes that returned a mixing percentage of 12.2%, 18.3%, and 4.9% lower than the maximum, but with 50.2%, 64.7%, and 23.7% fewer pressure drops, respectively. The approach combining statistical, numerical, and optimization to CDM allowed the finding of the best geometric design, which is mandatory for passive micromixers.
Keywords
Multi-Objective Optimization, Micromixer, Computational fluid dynamics (CFD), Genetic Algorithms, Constructal Design
