Research Projects

Understanding the interaction between turbulent flows and flexible structures is vital to address key performance and reliability challenges in wind energy systems. Wind turbines are particularly prone to structural fatigue, vortex-induced vibrations (VIV), and noise due to unsteady aerodynamic loading from turbulent inflow. Phenomena such as lock-in and flutter emerge from complex interactions between the unsteady flow and the structural modes, especially under high Reynolds number conditions. However, most numerical studies either neglect fluid–structure coupling or assume laminar flow, limiting their ability to predict nonlinear structural responses or noise in realistic environments. This project aims to conduct a comprehensive numerical study of turbulent flow-induced vibrations in flexible structures, focusing on wind energy applications. It will begin with unsteady Reynolds-averaged Navier–Stokes (URANS) simulations coupled with geometrically nonlinear structural models to establish a computationally efficient FSI baseline. Large eddy simulations (LES) will then be performed—first without structural coupling to identify turbulent features such as coherent structures and broadband fluctuations, and later with full FSI to capture feedback between wake and structural motion. The study will also investigate thermal effects where relevant and mechanisms of aeroacoustic noise generation due to unsteady structural oscillations interacting with turbulent vortices. The goal is to analyze the evolution of flow-induced instabilities across different flow and material regimes. Two configurations will be examined: (i) a cantilever mounted perpendicular to the flow (mimicking turbine towers/blades), and (ii) one aligned parallel (representing flexible blades). These models will provide insight into instability mechanisms and energy exchange processes between turbulent vortices and flexible structures. Building on prior experience in laminar FSI and thermally coupled simulations using the Stanford University Unstructured (SU2) multiphysics suite, this study will use SU2 for all simulations. It will focus on identifying turbulent wake features responsible for different vibration regimes under varied conditions. The project also seeks to develop surrogate models for dominant vibration modes and modal energy using machine learning and reduced-order modeling, based on key flow and structural parameters. These will allow rapid prediction of dynamics and critical transitions in future applications. In addition to wind energy, the developed framework will also apply to flow sensors, biomimetic structures, and unsteady heat transfer systems. By resolving turbulent dynamics and capturing nonlinear FSI effects, this work will advance the scientific understanding and computational modeling of turbulent flow-induced vibration phenomena in flexible systems.