Boeing uses Frontier supercomputer to solve turbulence physics

Understanding chaotic airflow remains a central obstacle in creating safer, faster, and more fuel-efficient aircraft. To address this complex physical challenge, Boeing researchers turned to Frontier, the world’s first exascale supercomputer located at the U.S. Department of Energy’s Oak Ridge National Laboratory (ORNL).

The team employed this powerful system to model how microscopic turbulent airflow and ice accumulation impact aircraft performance during critical flight phases, such as landing. These simulations are designed to assist engineers in refining aircraft designs, clarifying results that have previously confused wind tunnel experiments, and speeding up the development of airplanes with better fuel economy.

Impact of aerodynamic drag reduction

Even minor adjustments in aerodynamics can yield significant operational benefits. Research from the airline indicates that reducing aerodynamic drag by just 1 percent during cruise flight can lower fuel consumption by 0.75 percent. This efficiency gain has the potential to decrease operating costs for airlines and possibly lead to lower ticket prices for passengers.

Airflow around an aircraft frequently shifts between smooth, orderly motion and chaotic turbulence. Predicting the exact timing and manner of this transition continues to be one of the most difficult unsolved problems in fluid mechanics. The complexity arises from the vast number of air molecules interacting near the aircraft’s surface.

Researchers have traditionally focused on the boundary layer, the thin sheet of air flowing closest to a wing, where minute changes in airflow dictate the amount of drag experienced by the aircraft. Capturing these interactions requires simulations that analyze airflow at nearly microscopic levels. However, increasing the detail included in these models makes the calculations significantly more computationally demanding.

Limitations of traditional CFD methods

Aircraft aerodynamics are typically investigated using a combination of wind tunnels, flight tests, and computational fluid dynamics (CFD) simulations. While wind tunnel testing and flight experiments provide valuable data, they are often expensive and time-consuming. CFD offers a faster alternative by dividing an aircraft into a digital mesh composed of hundreds of millions of tiny cells, some measuring only millimeters across.

The software calculates properties such as pressure, velocity, and temperature for every cell, then integrates them to form a complete picture of the airflow surrounding the aircraft. Most engineers rely on Reynolds-averaged Navier-Stokes (RANS) simulations, which estimate turbulence rather than calculating every swirl of air. Although this method is fast enough for full aircraft simulations, it cannot accurately capture rapidly changing turbulent flows.

Higher-fidelity approaches offer much greater accuracy but require vastly more computing power. This is where Frontier made a decisive difference. Capable of performing 2 quintillion calculations per second, the exascale supercomputer enabled Boeing to run simulations that would otherwise take months on conventional industrial systems.

Exascale computing accelerates research

Researchers used the system to study how different ice shapes and textures influence wing performance and to model turbulence in far greater detail than previously possible. The simulations also helped evaluate aircraft anti-icing systems and explained observations from wind tunnel tests that had remained unexplained.

According to Boeing, the higher-fidelity models predicted an aircraft’s lift coefficient to within 1 percent, giving engineers greater confidence to move promising designs toward production. The team believes these findings could benefit the wider aerospace industry, as the simulation methods are based on a widely studied aircraft model that other researchers can build upon to improve future aircraft designs.