Faculty Advisor

David Olinger

Faculty Advisor

John Blandino

Faculty Advisor

Nikolaos Gatsonis

Faculty Advisor

John Sullivan

Identifier

etd-051110-094009

Abstract

Wingtip vortices are an important problem in aerodynamic and hydrodynamic engineering because of their contribution to induced drag, tip cavitation, and wake turbulence. These effects decrease equipment efficiency and lifespan, which increases application costs. Biology provides an inspiring solution to this problem in avian flight through the spreading of primary feathers. Previous studies have shown increased lift to drag ratio and efficiency of wings and propeller blades through modified wingtip geometry. The goal of this project is to optimize the tip geometry (primary feather angle) of a test wing for minimal tip vortex strength using genetic algorithms to mimic natural design evolution. Ultrasonic transducers are used to measure the wing tip vortex circulation in wind tunnel tests for each candidate design. Although neither angle of attack series converged completely, there was partial convergence in each. Due to the fluctuations in the low angle of attack tests, the parent selection algorithm was altered for the high angle of attack series, which resulted in improved convergence trends. A genetic algorithm that used uniform crossover breeding, a 20% mutation rate, and roulette wheel parent selection methods was used to generate an improved tip geometry at a low angle of attack of 6° and a freestream velocity of 15.25 m/s over the course of 17 generations. This improved design consisted of three key features, a staggered leading edge, a drastic mid-section vertical separation, and an upswept trailing edge. A second algorithm, which employed uniform crossover, a 20% mutation rate, and an elitist selection roulette parent selection, provided an improved tip geometry for a 12° angle of attack at a freestream velocity of 11.5 m/s. This improved design consisted of three key features, a downswept leading edge, a drastic mid-section vertical separation, and an upturned trailing edge. Both results showed that the wing tip vortex strength can be reduced by approximately 20% by manipulating tip geometry and that the trailing edge traits produce the most prominent effects on vortex strength.

Publisher

Worcester Polytechnic Institute

Degree Name

MS

Department

Mechanical Engineering

Project Type

Thesis

Date Accepted

2010-05-11

Accessibility

Unrestricted

Subjects

genetic algorithms, optimization, tip vortices, winglet, vortex strength

Share

COinS