| Aeroacoustics analysis of turbofan inlets |
| Thursday, 06 September 2007 05:35 |
|
Officially, acoustics is the science of sound creation and transmission, but in many ways, acoustics is more like an art or engineering. The creation of sound involves a complex interaction of shapes, materials, and energy. We are far from having reliable formulas that can capture the behavior of real-life structures. That's one of the reasons the New York Philharmonic is moving out of the modern, but acoustically unsatisfying, Avery Fisher Hall, and back to the hundred year old Carnegie Hall whose structure adds a glowing warmth and richness to performances. Often the only way to improve the acoustic behavior of a structure is through trial and error. Of course, it's not really possible to build a symphony hall over and over til you get it right. But these days, you can almost do that on a powerful computer, and get reliable results. The art of acoustic modeling requires the careful study and simulation of the materials, shapes, interactions and stresses of a complex physical system. A well designed model will allow an engineer to tinker with any part of the system. Especially when landing, most of the noise from an aircract comes from the rotating turbomachinery blades, rather than the engine, which is slowing down. One of the peculiar properties of this noise is that it is "discrete", that is, it is concentrated in certain frequencies, almost like a musical instrument in which a few keys have been pressed. Conventional methods of noise suppression are better at handling broad band noise, that is spread out evenly over many frequencies. The things that engineers would like to examine include changes to the shapes of the fan inlet, the fuselage and the wing. It's really not possible to do experiments on a jet engine in flight, and even doing wind-tunnel experiments is costly, and not completely realistic. You can fit the engine in, but the acoustic behavior will not be correctly modeled because the engine, the wing and the entire aircraft interact to create the sound. Unfortunately, the roar of the engine will actually get the wind tunnel itself vibrating, and this will change the overall results as well. Thus, by conducting experiments through computer simulations, Hussaini and his research group can reduce the cost of traditional experiments, and find good solutions to the problem of aircraft noise. M. Y. Hussaini, Dan Stanescu & Jinquan Xu |
Acoustic modeling can also help when there are sounds you don't want to hear. The problem of suppressing undesirable sound is familiar to anyone who's flown in an aircraft. But the biggest complaints come from people who live on the ground, near airports. Noise complaints kept the Concorde from flying anywhere but over the ocean; new noise regulations enacted in Europe mean that aircraft designers must control the noise emitted from their planes, or lose their markets. 
Of course, none of these problems occur in a computer model. With sufficient care, it is possible to simulate the behavior of an entire aircraft in flight over any kind of landscape. And if necessary, any part of the aircraft can be reshaped by changing a few lines of computer code, rather than by cutting and welding steel. Computer simulation can go through vast numbers of experiments, leaving wind tunnel tests to the end, when the most suitable candidates have been identified. 