"Output-Based Adaptive RANS Solutions on a Multi-Element Airfoil" by Carmen Ursachi
ABSTRACT: This talk studies the effect of meshes on RANS solutions for a 2D multi-element airfoil. Two types of meshes are considered: output-based adapted meshes and manually generated “best practice” meshes. Solutions are computed using several finite element and finite volume solvers. The convergence of aerodynamic coefficients, surface pressure, and skin friction on these meshes are studied in order to evaluate the accuracy and cost of the solution. At similar node counts, the adapted meshes generally result in more than an order of magnitude reduction in lift and drag error compared to the manually generated meshes. In addition, the characteristics of the adapted meshes are investigated at a prescribed error level for several different finite element discretization orders. This analysis provides insight into how the discretization order affects the mesh, along with the resulting solution accuracy and cost.
BIO: Carmen Ursachi is a PhD student working with Prof. David Darmofal. Her research interests are in computational fluid dynamics and mesh adaptation for aerospace problems.
-------------------------------------------------------------------------------------
"Closure Modeling for Three-dimensional Integral Boundary Layer (IBL3)" by Shun Zhang
ABSTRACT: We will introduce the development of a fully three-dimensional (3D) integral boundary layer (IBL) method that is intended for effective and efficient viscous aerodynamic analysis. This talk will specifically focus on the closure modeling for 3D IBL. A data-driven and physics-constrained approach is adopted in constructing regression-based closure models. Also, the idea of field inversion is experimented with for calibrating the turbulence model in the IBL context.
BIO: Shun Zhang is a PhD student advised by Prof. Mark Drela and also works in the SANS group led by Prof. David Darmofal. His research interests are broadly in computational methods for partial differential equations and machine learning for physical modeling, especially for aerodynamic applications. Shun has also grown a passion for education, particularly thanks to serial TA experiences for undergraduate fluid dynamics courses at MIT and the nonstop parenting of an infant at home.