Helicopters are noisy, they shake, and they are fickle in gusty winds. Better computational tools can help them fly more safely and smoothly and carry more cargo.

Simulating the aerodynamics of an airplane wing, with the unpredictable whorls and eddies of turbulent air flying past, pushes the most powerful supercomputers to their limits. Helicopter rotors present an even more difficult modeling challenge.

Wings, after all, are firmly fixed to the plane, and there are generally just two of them. Helicopter rotors, on the other hand, are essentially a bunch of wings whirling hundreds of times a minute, amplifying the vibrations, bends, and twists that occur in flight. When studying aerodynamics, the interplay between the vibrating aircraft and the air around it makes for a complex mathematical tangle.


  Gopal Gaonkar Partly because of the inherent complexities, partly because they didn't receive much attention until the Vietnam War, helicopters lag far behind airplanes in terms of performance and reliability. Helicopters are noisier, more difficult to fly, and give a bumpier ride than conventional aircraft.

"A helicopter is an intrinsically more complicated body compared to a plane," says Gopal Gaonkar, a mechanical engineering professor at Florida Atlantic University in Boca Raton.

Gaonkar is taking advantage of the IBM SP2 supercomputer at the Maui High Performance Computing Center, a member of the Alliance's Partnership for Advanced Computational Services, to improve the design of helicopters using Floquet analysis, the primary mathematical tool for helicopter stability investigations. Floquet analysis isn't new -- it's about 125 years old. But, like other aerodynamics calculations, it takes a heap of memory and computation time, making it impractical for all but the simplest designs.


Access Online | Posted 2-15-2000