Date of Award

Winter 12-12-2020

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Master of Engineering (ME)

Department

Physics

Advisor

Samuel Hess

Second Committee Member

Wilhelm Friess

Third Committee Member

Richard Eason

Abstract

Mars exploration and UAV development have both advanced significantly over the past century, and are now being considered in tandem. Currently needed are UAV propellers that can operate in the Martian atmosphere. Flow will be in the range of Re < 20,000, creating extreme conditions not typically examined. A Blade Element Momentum Theory (BEMT) algorithm is developed using a variety of corrections designed specifically for low Reynolds number and rotational flows. Due to both the simplicity of the basic BEMT formulation, corrections are easy to put in place and often necessary to achieve accurate estimates. Aerodynamic coefficients are determined from XFOIL code, and have questionable accuracy in this regime. To account for this, a correction model is developed by comparing XFOIL results to experimental results of airfoils at low Re. This is all tested against a previous low Re propeller experiment. The results of this comparison are used to adjust the values in the correction, to produce more accurate results for theoretical design.

From here, a design philosophy for the propeller is developed using established methods and previous experimental data. High thrust is prioritized, with efficiency being a secondary concern. A hard mach limit of 0.7 is set to avoid major drag penalties, limiting the usable ranges of RPM and radius. Airfoil designs are then examined, based on previous designs, theoretical intuition, and experimental data. A modified version of the S1223 airfoil is adopted for its favorably high Clmax and high stall angle. From here, optimization can be used to determine the final dimensions of the propeller. The BEMT algorithm is used to create a broad set of data, over a range of design variables, which is then fitted to thrust and efficiency functions using non-linear regression. A Non-Dominated Sorting Genetic Algorithm (NSGA) is well suited to optimizing multiple objective functions with multiple design variables, and thus is adopted to optimize the design. The results of the optimization confirm previously published theories, and produce three possible propeller designs, a high thrust model, a high efficiency model, and a compromise between the two. These designs were then modeled, meshed and simulated using the ANSYS Fluent software suite. BEMT thrust estimates were found to be within an average absolute error of ~41% from the simulated results, while moment was within an average absolute error of ~104%. This discrepancy can likely be attributed to the inaccurate drag data being sent into the BEMT algorithm, and the lack of a method to correct said data. With a procedure established for design and testing, new propellers can be created and verified, likely with improvements in accuracy from the initial estimates.

Download

Share