Computational Fluid Dynamics (CFD) is an important and widely used method for analyzing the aerodynamic performance of various geometries, the results from which are crucial to aircraft design and development. However, using CFD results in two different kinds of errors: numerical errors (associated with the discretization scheme and point distribution of the mesh) and modeling errors (associated with the models used to simplify the problems such that they can be solved at a reasonable computational cost). The goal of this project is to compare the effect of using different turbulence models to simulate transonic regimes. To do so, a method called mesh adaptation will be used to drive numerical errors to machine zero when performing CFD simulations, thus isolating the modeling error and allowing for rigorous comparisons of different turbulence models. The project will focus specifically on simulating 2D and 3D geometries under transonic conditions at various angles of attacks, as these configurations are commonly used in aircraft designs. It is important to have a thorough analysis and comparison of turbulence models in these configurations because different turbulence models predict different shock locations and output functional values (e.g. lift, drag, and moment) in transonic regimes; in a field where even small inaccuracies in prediction can come at a significant cost, it is crucial to determine which turbulence model is “best” to use.

I am excited to participate in the SuperUROP program as it will help me hone my research and technical communication skills while building on a project I’ve been working on at MIT for the past year. This year, I look forward to deepening my understanding of CFD, expanding my computational skillset, and gaining a more formal introduction to research through this program.