Date of Award

Winter 2024

Project Type

Thesis

Program or Major

Mechanical Engineering

Degree Name

Master of Science

First Advisor

May-Win MT Thein

Second Advisor

Nathan NL Laxague

Third Advisor

Se Young SY Yoon

Abstract

This thesis investigates the feasibility of using Uncrewed Aerial Vehicles (UAVs) and existing state estimation techniques for detecting aerodynamic wakes around bluff bodies. Aerodynamic wakes are complex, turbulent flow regions created by obstacles. Accurately measuring these wakes is crucial for applications such as optimizing wind energy production, improving aircraft safety, and enhancing atmospheric research. Traditional measurement methods, such as ground-based sensors or piloted aircraft, have significant limitations, including restricted spatial coverage and high operational costs. UAVs, with their versatility and cost-effectiveness, present an alternative approach to measuring wakes in challenging environments.

This research focuses on three state estimation techniques: the Extended Kalman Filter (EKF), the Sliding Mode Observer (SMO), and the Extended State Observer (ESO). The performance of these estimators is evaluated through a combination of Computational Fluid Dynamics (CFD) simulations and experimental UAV flights. The CFD simulations, based on Large Eddy Simulation (LES) methodology, generates a realistic wake environment around a backward-facing step, representing the bluff body. The simulations provides a detailed wind field, which is used as a benchmark to assess the accuracy of the state estimators. Experimental testing involves both laboratory and field tests using UAV platforms equipped with onboard sensors and anemometers.

The EKF effectively estimates wind trends while filtering out noise making it a promising wake sensor for experimental data. The SMO generates the least Root Mean Square Error (RMSE) compared to that of the other estimators while in high-frequency wind dynamics during simulations, showing robust performance under rapidly changing wind conditions. The ESO, despite its limitations in high-frequency conditions, provides a computationally efficient solution for estimating general wind behavior but is not well-suited for detecting wakes due to its lack of precise state estimates.

This study highlights the potential of UAVs equipped with state estimation algorithms for autonomous wake sensing, offering a scalable and flexible approach for applications in wind energy, environmental monitoring, and airspace management. The findings indicate that each estimation technique has unique strengths, making them suitable for different aspects of wake measurement. Future research includes improving UAV dynamic models, mitigating sensor noise, and exploring the use of multiple UAVs to enhance the real-time spatial resolution of wake measurements.

Download

Share

COinS