About the SA Blog Network



Opinion, arguments & analyses from the editors of Scientific American
Observations HomeAboutContact

Predicting small-scale turbulent flows could lead to more efficient airliners and ships

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

Airliner in flightIt is a phenomenon that has eluded description for centuries. Today’s supercomputers are not up to the task of simulating it in detail. And the great physicist Richard Feynman reportedly called it "the most important unsolved problem of classical physics."

What grand conundrum could possibly cause such trouble? The answer is surprisingly mundane: turbulent fluid flows. When a solid object passes through air or water, the interface becomes a chaotic region of almost innumerable fluctuations and eddies as the surrounding medium passes over and around the solid object.

It is those chaotic flows and the drag they generate that account for up to half of the jet fuel burned during an airplane flight, and an even higher share of fuel usage on large ships and submarines, according to a study in the July 9 issue of Science. But modeling such complex behavior in realistic detail is a task far too daunting for today’s technology.

The effects of turbulence near a surface, whether the body of an airplane or the side of a ship, are difficult to predict, the study’s authors explain. "The most important processes occur very close to the solid boundary—the region where accurate measurements and simulations are most challenging," report Ivan Marusic, Romain Mathis and Nick Hutchins of the department of mechanical engineering at the University of Melbourne in Australia.

The Australian group has produced a new model that predicts some properties of such fine-grained "near-wall" turbulence from large-scale effects measured farther away from the surface. The outer layers of fluid flow, the authors found, have a twofold impact on the near-wall turbulence: the outer turbulent "superstructures" both deliver energy to the wall region and modulate the scale of the fluctuations there. The new model, which incorporates those effects, was found to match well with results from a simplified experimental setting with a 27-meter flat surface in a wind tunnel.

In an accompanying commentary in Science, Ronald Adrian of Arizona State University notes that further work will be needed to verify that the model works for all interfaces, not just the simplified wind-tunnel surface, and that it can predict the full complexity of turbulence. If models such as that developed by the Australian team could bring Feynman’s great problem one step closer to being solved, engineers might be able to better account for its effects by designing surfaces that experience less drag. "The ability to predict such behavior," the study’s authors write, "often implies opportunity for control."

Photo: © iStockphoto/sculpies


Rights & Permissions

Comments 5 Comments

Add Comment
  1. 1. quincykim 8:00 pm 07/8/2010

    Sounds promising, and with far-reaching impact if successful. I wish them all the best.

    Link to this
  2. 2. sanoran 3:23 am 07/12/2010

    Just watch a fish or bird move, and you will know that our planes and ships are clumsy. But energy has to be a bit more expensive for such research to be serious. For example, GM and Ford still spend more effort on body styling than on combustion chamber modeling, …. because as long as energy is cheap, they don’t care. Cheap energy is coming to an end, and the airlines are the first to take notice. But energy is still to cheap for efficiency to matter.

    Link to this
  3. 3. kristi276 10:52 pm 07/14/2010

    As I understand, as a lay person, the basic design of ships and airplanes is based on the arrow principle where an object is able to cut through a medium with a less drag efficiency. Is the arrow design the best overall efficient means to accomplish this, maybe instead of trying to get the environment to suit the design why not get the design to suit the environment to lessen the overall surface friction and resistance.

    Are supercomputers better than GP U’s for scientific computing s? I have heard that GP U’s are faster with higher computational speeds, maybe instead of using Supercomputers for creating these s, how would an array of GP U’s with high end architectural graphic cards work?

    Link to this
  4. 4. Quinn the Eskimo 2:09 am 07/15/2010

    Maybe we could "grid" our iPhones?

    There’s and App for that.

    Link to this
  5. 5. tichead 12:15 am 07/24/2010

    Please forgive my ignorance, but what is a GP U?

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article