Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
DESIGN OPTIMIZATION OF SOURCES IN FLUID FLOW SYSTEMS USING COMPUTATIONAL FLUID DYNAMICS-BASED ADJOINT METHOD
Submission Author:
Javier Aliaga Rivera , RJ , Brazil
Co-Authors:
Javier Aliaga Rivera, Marcos Sebastião de Paula Gomes
Presenter: Javier Aliaga Rivera
doi://10.26678/ABCM.COBEM2023.COB2023-2194
Abstract
In this study, we develop a method for design optimization of a discrete number of sources in fluid flow systems using computational fluid dynamics-based adjoint method. The optimal location and magnitude of discrete sources in the domain, to achieve a determined steady-state condition, is the focus of our research. To this end, we present the derivation of the sensitivities with respect to the coordinates and magnitude of the sources, using the continuous adjoint method. The comparison of the calculated sensitivities with the ones obtained from the Finite Difference Method proves the accuracy and robustness of the proposed method, allowing the systematic evaluation of such sensitivities without the need to resort to automatic differentiation or the expense associated with finite difference approximations. The principal motivation for this work comes from problems in optimal design of valves and flow control. To illustrate the method, standard scalar transport in steady-state is first presented in one dimension, followed by an extension to two dimensions. Furthermore, a number of numerical studies are carried out for different configurations and operating conditions. The different cases demonstrate the method's capability by applying it to the optimal design of location and magnitude of distributed sources with little computational effort. Finally, possible applications and limitations of the methodology are discussed.
Keywords
continuous adjoint, source optimization, Parametric Optimization, CFD
