RESEARCH PERSONALITY
Flying High
By Lea McLees
IF YOU OBSERVED the wings of the airplane you last flew in, you probably noticed that they moved slightly during your stay in the sky. That movement, although important, hasn't always been taken into account in the study of fluid -- such as air -- flowing over objects -- such as airplane wings.
A code that incorporates the structural and aerodynamic
equations of motion and computational fluid dynamics provides the most
accurate picture to date of how air behaves around the body of a plane,
says Dr. Marilyn Smith.
|
But Dr. Marilyn Smith of the Georgia Tech Research Institute (GTRI) is trying to change that. She, her colleagues and sponsors at the Air Force's Wright Laboratory Aeromechanics and Aeroelasticity Divisions are integrating computational fluid dynamics (CFD) applications with the structural equations of motion. They apply these combined sets of equations to form the basis of aeroelasticity -- the movement of a plane's structure during flight, using realistic air forces.
By solving these equations on powerful computers, Smith is making important progress in methodologies that are enhancing aircraft design, and making airplanes and helicopters less susceptible to damage from fatigue.
"We've had calls from all over the country wanting to access the methods and the research that we've done, so I think it may be making an impact," says Smith, a senior research engineer.
Smith's work is centered in the CFD world of predicting drag and visualizing flow. A relatively new tool, CFD is based on differential equations that, when manipulated on a computer, can help researchers predict fluid flow behavior.
Smith is integrating CFD applications with the structural equations of motion. She applies this combination of equations to the study of aeroelasticity.
"Most CFD codes today only consider a rigid body," says Smith, who works in GTRI's Aerospace and Transportation Laboratory. "But that's not what really happens in flight in an aircraft. In flight, if you're sitting near the wings and you look out the window, you'll see the wings move. The structural equations of motion address that phenomenon, along with the air loads generated by the flight speed. So, by incorporating the structural and aerodynamic equations of motion and CFD in the same computer code, you can get the coupling and the interaction between structure and fluid flow -- and that's what really happens in flight."
The combination provides the most accurate picture to date of where air flows smoothly over a plane's body and where it whips into turbulent whirlpools, or vortices, that vibrate or buffet parts of the vehicle. These vibrations fatigue parts and may cause them to wear out more quickly than they would otherwise, Smith says.
"By performing computational aeroelasticity at such a high level, we can predict changes in performance that result from design modifications or fatigue," she explains. "We can simulate a situation numerically, see what's happening in the flow field, and design better parts."
Her biggest contribution has been making the study of aeroelasticity, using any given CFD computer code, easy.
"Different CFD codes work well for different applications, different aircraft components or flight regimes," Smith explains. "Right now we're concentrating on static aeroelasticity, which will give us a good idea of the performance changes and the overall flow field changes. Dynamic aeroelasticity, which is the prediction of flutter points -- the points at which the structure begins to be become dynamically unstable -- is probably two to three years away."
Smith collaborates with several Georgia Tech colleagues: Dr. Dewey Hodges, professor of structural dynamics in Tech's School of Aerospace Engineering; Dr. Carlos Selznik, a post-doctoral aerospace student in structures; Dr. Lakshmi Sankar, professor of fluid dynamics in the School of Aerospace Engineering; and Dr. Olivier Bauchau, professor of structures in the School of Aerospace Engineering.
These researchers make Georgia Tech one of five institutions in the United States that combines high-level aerodynamics and structures research. The other locations are NASA's Langley Research Center in Virginia, Ames Research Center in California, the Air Forces' Wright Laboratory, the Lockheed-Martin Engineering Services Co. and Skunk Works.
"There are other people working in aeroelasticity all over the country, but not to the complexity that we are working on it in aircraft or in aerodynamics," Smith says.
Computers have only recently become fast enough to do the work Smith and her colleagues perform. They develop codes in C and FORTRAN and test them using a Cray supercomputer.
"We can tell the impact of the structure motion by comparing the pressures on the vehicle with the rigid vehicle results," Smith explains. "The pressure are integrated into loads which are the forcing functions for the structural equations of motion. We also look at the flowfields surrounding the aircraft, particularly for vortex movement. These data are examined by both conventional plots and by color contouring graphics."
As she has incorporated structural dynamics knowledge with Euler/Navier-Stokes equations of fluid motion around aerodynamic vehicles, Smith has learned some interesting things.
"It's interesting to see how the flow field changes," she says. "Obviously things move -- vortices will change their directions and strengths, for example. You get differences in performance of the aircraft, or of the component you're looking at."
The implications of her findings are important.
"We've been called upon to help fix problems in aircraft," Smith says. "The classic case is the F-15 vertical tail buffet problem. By incorporating aeroelasticity into our CFD analyses early, we can predict more accurately where these flow field phenomena will appear on the body. If we know where the vortex or unsteady flow is going, we can either design devices to turn the unsteady flow away from that component, to prevent the damage -- or we can beef up the structure enough that it will not move with the unsteady flow around it.
"If we can simulate a situation numerically and see what's actually happening in the flow field, we can design a better part," she explains. "It won't be a hit or miss process anymore."
The biggest challenge for Smith so far? Maintaining the integrity of the numerical solution she's using when linking motion to a flexible surface. In addition, nodes -- the points on the plane at which she tests the equations -- have to be placed correctly, especially when running tests for a moving vehicle. When a vehicle isn't still, vortices move and change directions.
"If you don't have the nodes in the right places, if they're not smoothly distributed, or if there are not enough in a certain area where you have a lot of flow fields, then you won't get as accurate of an answer as you would if they were," Smith explains.
Smith has enjoyed a lifelong fascination with flight. She grew up watching Apollo missions on television and planning to explore space. The space program hit a lull during her high school years -- but during a college co-op job at NASA Langley Research Center, airplanes piqued her flight fascination.
"I like the mathematics of it and the physics of it," she says. "I like the ability to see what's going on around the aircraft."
Smith worked as an associate scientist at Lockheed Aeronautical Systems Co. in Georgia, and as a senior flight test engineer at Lockheed-Georgia Co., and as a member of the technical staff at McDonnell-Douglas Helicopter Co. in Arizona. She joined GTRI in 1990.
"I've had a wide range of practical engineering experience -- all the way from flight testing to design and wind tunnel testing," she says. "I have a practical outlook on things."
While working full-time in industry she earned master's and doctoral degrees in aerospace engineering from Georgia Tech. Smith was the sixth woman to graduate with a Ph.D. in aerospace engineering from the school.
Her master's degree required two years to complete. Her Ph.D. required nine years from starting classes to defending her dissertation. During those nine years she moved across the country twice, changed jobs twice, and had a daughter.
Smith is a senior member of the American Institute of Aeronautics and Astronautics, an associate member of the AIAA Fluid Dynamics Technical Committee, and a member of the American Helicopter Society. She was a recipient of the 1982 Best Woman Engineering Student and Best Cooperative Student awards from Atlanta's chapter of the Society of Women Engineers.
She occasionally teaches high-speed aerodynamics, hypersonics or transonic aerodynamics at her alma mater, encouraging students to develop a broad foundation of knowledge -- even if they are enamored of one particular facet of aerospace engineering.
"They learn that as design people, for example, they will use numerical simulation and they need to know the theory behind it...and that the structures people need to understand aerodynamics, and how it impacts on structures," she says. "And I try to introduce a balance of good, practical engineering applications in every topic that I teach. There's no greater satisfaction than to be able to teach a course in aerodynamics to the younger generation and be able to introduce practical aspects -- so that a new generation of engineers sees the worth of it while they're still in school."
Smith wants to use the integrated codes she and her colleagues have worked on to design new aircraft. She's also interested in incorporating an additional discipline, acoustics, into her current work.
"I'd like to determine how these vibrations and motions affect the sound environment," she says. "Sound plays a very big role in determining vulnerability and survivability aspects of military aircraft."
Sound also is important in the commercial sector. Bad acoustic signatures from an airplane can cause noise problems for those living near airports.
"It's not only important to design aircraft and aircraft parts that are less susceptible to damage from fatigue -- we can use the results from the codes we hope to develop to make their takeoffs and landings more palatable to the ears of the people on the ground," Smith says.
Further information is available from Dr. Marilyn Smith,
Aerospace and Transportation Laboratory, Georgia Tech Research
Institute, Georgia Institute of Technology, Atlanta, GA 30332-0840.
(Telephone: 770/528- 7804)
Send questions and comments regarding these pages to Webmaster@gtri.gatech.edu
Last updated: May 30, 1997