Bakhtian an Astronaut in the Making
By Gabriel Chen, April 2005 Most kids discard dreams of becoming an astronaut as fast as they abandon childhood toys, but for senior NoÃ«l BakhtianÂ– a mechanical engineering and physics double major at DukeÂ– the dream never let go.
Some four years ago, Bakhtian, then a senior in high school, was on a flight back from a visit to MIT when she fell into conversation with a fellow colleague of Bakhtians dad who happened to be on the same plane. He convinced her to apply to Duke where he had completed medical school. He told Bakhtian that she would get a well-rounded education there.
Bakhtian said that when she visited Duke for the first time she was immediately drawn to the Chapel, which stands at the heart of the Universitys West Campus. As one of the largest Gothic-style churches ever built, with three world-class organs and "mesmerizing" stained glass windows, Bakhtian fell in love with it straight away. Once enrolled at Duke, Bakhtian couldnt help but spend time in the chapel. She became one of the weeknight chapel attendants, a member of the chapel choir, and made friends with the chapel staff. "The Chapel is an oasis of peace in the middle of my crazy, fun, and sometimes stressful Duke existence, and the people there are my family."
Realizing early on her affinity to all things related to airplanes, space, and NASA, Bakhtian has become involved in activities and studies related to these interests. Studying abroad for a semester at the University of Sydney in Australia, Bakhtian enrolled in an aircraft design and construction course in which she actually built a 2-seater Jabiru single-engine kit airplane that flew and was eventually sold. "That hands-on experience was invaluable in that it made me realize how knowledge of the actual construction procedures and methods fortifies the concept development and design processes," she said.
Last semester, she enrolled in a graduate level Aerodynamics course where she learned the fundamentals of sub- and supersonic fluid flow applied to wings and bodies. Bakhtian is also a Pratt Fellow researching nonlinear aeroelasticity with guidance from Dr. Earl Dowell, Dukes William Holland Hall Professor of Mechanical Engineering and Materials Science and Dr. Peter Attar, an NRC Associate Researcher at Wright-Patterson Air Force Base.
Aeroelasticity is the study of the dynamic interaction between an aerodynamic flow and an elastic structure, such as aircraft wings in high speed flight. One critical aeroelastic phenomenon is flutter, a dynamic instability of the system which causes a growing oscillation of the surface leading to large amplitudes and stresses which can cause failure.
"Sometimes," Bakhtian states, "nonlinearities in the system limit the oscillation amplitudes. This is called Limit Cycle Oscillations, or LCOs." Specifically, Bakhtians research, funded by the Air Force Office of Scientific Research (AFOSR), aims at testing how well a high fidelity aeroelastic computational model of this phenomenon correlates with in-house experimental data.
"Pretend youre an Air Force pilot," Bakhtian said. "Youre out cruising around in your F-16 and decide to increase the speed, when you feel that the wing is going into flutter. A normal reaction would be to decrease the planes speed to what it had been previously, but even as you do this, you realize that the wings wont stop fluttering. This phenomenon, called hysteresis, simply means that it is possible for the wing to be fluttering below the flutter velocity. The fighter pilot needs to know how much more to decrease his speed, so as to stop the flutter."
Test flights to determine this data are costly and dangerous, so Bakhtians group is building a set of computation tools that solve for critical values such as flutter velocity, the limit cycle oscillation (LCO) amplitudes, and the hysteresis effects, all without putting a plane in the air.
Bakhtian said that the computational model she has tested, a nonlinear finite element code (ANSYS) coupled with an aerodynamic vortex lattice code (simplified through use of the reduced order method of Proper Orthogonal Decomposition), has predicted both the flutter speed and flutter frequency accurately. However, the correlation between theory and experiment has not been as accurate for the prediction of LCO magnitudes. Also, while hysteresis with respect to flow velocity has been noted in the experiment, this phenomenon has not been noted in the simulation results.
"We wanted our computer model to have better correlation with the wind tunnel results," Bakhtian said. "The simulations gave accurate flutter velocities, but we werent seeing hysteresis and the LCO amplitudes were a bit low. We think that in the experiment there was flow separation off the edges of the plate which caused enhanced lift and thus higher LCO amplitudes. Separation is a nonlinear effect and could possibly have also caused the hysteresis we saw in the wind tunnel experiments."
Bakhtian further explains that the complexity of both the vortex lattice code and the separation phenomenon makes it difficult to integrate the two: "To model separation, you attach a wake on that edge, but often the wake will try to interact with the wing vortices [of the vortex lattice] which is why we hadnt included separation in our computation model."
Another consideration was the possible ambiguity resulting from the straight side edges. Bakhtian said that the flapping flag configuration for these tests (so named because their physical model flutters like a flag on a breezy day) was a thin aluminum sheet, rectangular in shape. When running simulations, it was not clear whether the side edges were defined as leading edges or trailing edges. To remedy this, the group designed a new flapping flag configuration that was trapezoidal rather than rectangular. The next step, according to Bakhtian, is to add separation into the computational model and rerun the simulations and the experiment in order to compare the new results. She has already started wind tunnel tests in Hudson Hall on four new trapezoidal models, a "fresh look at research after a year and a half of computer simulations!" and says she cant wait to see how these new results compare.
As for her post-graduation plans, Bakhtian has been awarded the prestigious Churchill Scholarship, an opportunity for one year of research at Churchill College of Cambridge University in England. There, Bakhtian will be studying vorticity and turbulence and experiencing life in Britain. Meanwhile, she has been offered admission to Stanford, Caltech, Duke, and several other prominent schools, one of which she will attend upon returning to the States. Several years down the road, she hopes to complete her Ph.D. in aeronautical engineering and then apply to the astronaut program.
"The thing Ill miss most about Duke are the interactions Ive had with the faculty and administrators," Bakhtian grinned. "When I think back over the last four years, of course I remember taking classes and the studying, but Ill never forget things like playing pool with Dr. Bliss, having involved discussions about flow separation over AIM, philosophical talks coming up during math office hours, discussing movies and books with our professors after class, analyzing clothing styles with Dr. Dowell, rearranging Dean Simmons office while she miraculously manipulated my schedule so I could graduate on time, popping into Dr. Nadeaus office to tell a random story, or just hanging out with MT and LB in the Deans office. Duke is an incredible place because of the people youre surrounded with daily- Get to know them. They have so much to share!"
Bakhtians hometown is Fort Myers, Florida. She graduated in May, 2005.