Thesis Defense - Lorenzo Fabris (MSME)

Lorenzo Fabris – M.Sc. in Mechanical Engineering  

Asst. Prof. Altuğ Melik Başol – Advisor

Date: 03.05.2024

Time: 12.30

Location: AB1 412

  MULTIPOINT ADJOINT SHAPE OPTIMIZATION OF A DIFFUSER VANE WITH SU2 OPEN-SOURCE SOFTWARE

Asst. Prof. Altuğ Melik Başol, Özyeğin University

Assoc. Prof. Dr. Özgur Ertunç, Özyeğin University

Assoc. Prof. Dr. Matteo Pini, Technische Universiteit Delft

Abstract:

Due to their capability to manage low and moderate mass flow rates to achieve high pressure ratios, centrifugal compressors are widely used in multiple industrial sectors. Nowadays, the design of turbomachines heavily relies on optimization utilizing computational fluid dynamics. However, optimization must be defined accurately and logically to be successful and productive.

The aim of this study is to present a viscous, multi-point, gradient-based shape optimization for a three-dimensional vaned diffuser flow scenario with the utilization of discrete adjoint method. The numerical analysis is carried out with the open-source computational fluid dynamics software SU2 (Stanford University Unstructured). The three-dimensional, steady, turbomachinery flow is initially analyzed with a previously, third part-implemented version of the SU2 Reynolds-Averaged Naver-Stokes solver, characterized by the utilization of non-reflective boundary conditions. Afterwards, the discrete adjoint solver automatically processes the numerical data of the flow field establishing the same convergence behavior as the primal solver itself, due to the automatic inheritance of its numerical properties.

Eventually, the numerical optimization led to a considerable decrease of losses as shown by analysis of total pressure loss, entropy and efficiency within the domain.

Bio:

Lorenzo Fabris received his B.Sc. in 2021 from Gugliemo Marconi University (Rome, Italy) – while developing his career in industry and his own company – with a thesis on biofuels utilization for aerodynamics engines utilizations. After leaving his occupation for the academic field, he continued his studies with a M.Sc. in Mechanical Engineering in Özyeğin University, specializing in Turbomachinery Computational Fluid Dynamics, numerical methods and gradient-based optimizations.