Record-breaking simulation of aircraft wake turbulence
Dangerous vortexes form when an aircraft takes off or lands. A combination of optimised numerical methods, a supercomputer and highly developed visualisation methods revealed how this process generates primary and secondary vortexes, and how the secondary vortexes cause the primary ones to decay.
Post-docs Philippe Chatelain and Michael Bergdorf, doctoral student Diego Rossinelli, Petros Koumoutsakos, Professor at the Institute of Computational Science of ETH Zurich, and Alessandro Curioni and Wanda Andreoni of the Department of Computational Sciences at the IBM Zurich Research Laboratory in Rüschlikon explain that they have succeeded in a “record-breaking simulation” of wake turbulence formed by the vortexes that occur when an aircraft lands or takes off. This involved the scientists using high-resolution calculations to simulate, with an accuracy two to three orders of magnitude higher than previously, the vortexes generated when an aircraft takes off or lands. Alessandro Curioni explains that the simulation also visibly revealed how the wake turbulence vortexes become unstable. He says this in turn allows a better understanding of the underlying physics and is thus useful in aircraft construction.
Modelling the air flow
The flow pattern was modelled by dividing the air up into very small elementary cells which were then simulated as particles independent of one another. Instead of studying the behaviour of several million particles as previously, the Zurich researchers have now succeeded with up to 6 billion particles. The scientists published their results in “Computer Methods in Applied Mechanics and Engineering”.
The vortex wakes generated by aircraft when they land or take off are dangerous: they can cause a following aircraft to crash. The special high-resolution simulation has now enabled the researchers to visualise for the first time how very small vortexes interact. This allowed them to achieve a large and previously unattained value for what is known as the Reynolds number. Petros Koumoutsakos explains that the larger this value, the more realistic is the simulation. He says this is because the number characterises the aircraft’s vortex wake and is proportional to the aircraft’s speed and wingspan. On the other hand the number of computational elements needed to simulate this turbulence increases as the square of the Reynolds number, thus limiting the possibility of simulating it by calculation. Koumoutsakos says that although the new simulation has allowed a closer approach to reality than ever before, the joint aim of the ETH Zurich researchers and IBM scientists is to be able to simulate even more realistic flight conditions.
Reducing environmental pollution
For their simulation the researchers used a supercomputer provided by IBM with 16,000 processors – corresponding to about 10,000 modern laptops. Through aircraft design adapted to it, the study can contribute to reducing the formation of vortexes and accelerating their self-destruction. Koumoutsakos says: “That is of the greatest importance for flight safety, noise reduction and air pollution.” Michael Bergdorf explains that “The structure of the vortexes can have a considerable effect on fuel consumption.” He says that a large proportion of aircraft noise is both generated by and carried by the vortexes.
Increasing flight frequency
Air traffic worldwide is growing by about five percent per year. Passenger numbers are forecast to double by 2025. This will lead to bottlenecks at airports. However, if a more exact knowledge of the generation of vortexes and their effect enables aircraft to be designed so that they form fewer vortexes and the latter destroy themselves more rapidly, this could also allow the prescribed separation distances between takeoff and landing times to be shortened, thus increasing flight frequency.
Chatelain, P. et al.: Billion vortex particle direct numerical simulations of aircraft wakes, Computer Methods in Applied Mechanics and Engineering 197, 1296-1304 (2008), doi:10.1016/j.cma.2007.11.016.
- 20.02.08: Impressive Results