High-Order FSI for Large-Deformation Aeroelastic Simulations

Student : Braden Frigoletto

Sponsor : Air Force Research Laboratory (AFRL)


Summary:

This research aims to develop an adaptive, high-order fluid-structure interaction (FSI) solver for large-deformation aeroelastic simulations. There are several objectives for this work. Firstly, we wish to demonstrate high-fidelity coupled FSI simulations. We plan to extend adjoint-weighted residual techniques to the coupled problem. We plan on estimating numerical errors for aeroelastic outputs of engineering interest. We wish to develop mesh adaptation algorithms for both fluid and structural discretizations. Using all of this, we plan to investigate the role of high-order formulations in improving the accuracy of FSI simulations. Validation of the simulation results must also be conducted and will involve wind tunnel experiments done within Michigan’s 5’-by-7’ wind tunnel. These experiments will utilize the pitch-and-plunge apparatus (PaPA) developed here at A2SRL. PaPA will allow models to fly with pitch and plunge degrees of freedom in free/free, locked/free, free/locked, and locked/locked configurations.

Braden’s responsibilities for this project will consist of handling the structures portion of the code within the FSI solver and helping to couple it with the fluids portion of the solver. Braden also assists in readying the instrumentation and data acquisition system for PaPA. The current FEM code selected for the project is the Toolkit for the Analysis of Composite Structures (TACS)[1]. TACS is a parallel finite element code for analysis and gradient-based design of advanced structures and was developed by Prof. Graeme Kennedy of Georgia Tech[2]. TACS will eventually be coupled with Xflow, a fluids solver developed by Prof. Fidkowski’s group[3] here at the University of Michigan. Currently, Braden is involved in benchmarking TACS to NASTRAN for various types of problems involving built-up structures and will begin developmental changes to TACS necessary to have it work with the chosen coupling scheme for the FSI problem.

[1] https://github.com/gjkennedy/tacs

[2] https://ae.gatech.edu/people/graeme-james-kennedy

[3] https://aero.engin.umich.edu/people/krzysztof-fidkowski/