Aeronautics

Simulating an Iced Wing for Aerodynamic Performance Predictions

Image from simulation ice flow over a wing
Snapshot of a wall-modeled large-eddy simulation of the flow over the outboard portion of a scaled Common Research Model wing with high-fidelity leading-edge ice. The image shows section slices of vorticity magnitude, where the darker color indicates regions of higher vorticity magnitude, and the lighter color indicates regions of lower vorticity. The inset highlights the intricate shape of the leading-edge ice. David Craig Penner, Gerrit-Daniel Stich, NASA/Ames

According to the Federal Aviation Administration, the National Transportation Safety Board revealed that between 2008 and 2021 there were at least 52 aircraft accidents and 64 fatalities attributed to structural icing during flight. Indeed, ice buildup on an aircraft poses a significant safety risk in part because it leads to reduced lift and increased drag. To help find ways to reduce the risk, NASA’s Advanced Air Transport Technology (AATT) and Transformational Tools and Technologies (TTT) projects are exploring state-of-the-art computational tools that can provide accurate and cost-effective aerodynamic predictions to help assess the risk for aircraft in iced flight. Predicting the aerodynamics of iced aircraft is particularly challenging due to the tendency of ice buildup to promote the development of large regions of highly turbulent and unsteady separated flow—even at low angles of attack. To better understand the aerodynamic prediction capabilities of scale-resolving numerical methods for iced flight and the fundamental flow physics involved, the Launch, Ascent, and Vehicle Aerodynamics (LAVA) group at NASA’s Ames Research Center performs scale-resolving computational fluid dynamics (CFD) simulations of an iced Common Research Model wing at several angles of attack using a hybrid body-fitted/immersed-boundary wall-modeled large-eddy simulation (WMLES) approach.

Quick Facts

Iced wing CFD simulations were enabled by the high-performance computing resources at the NASA Advanced Supercomputing facility, which allowed us to leverage state-of-the-art scale-resolving methods to provide aerodynamic predictions with the accuracy needed for aircraft design and certification.

David Craig Penner,
NASA Ames Research Center
  • Each unsteady case required approximately 400 million grid points and was run with the LAVA Curvilinear flow solver on 6,400 cores for approximately two days. The team used both the AMD Rome processors on NASA’s Aitken supercomputer and the Intel Skylake processors on the agency's Electra supercomputer to complete their work.
  • The hybrid body-fitted/immersed-boundary WMLES approach that was used to incorporate the ice shape into the simulations is an effective way to model the influence of ice on the flow, as it circumvents the need to mesh the ice directly using body-fitted grid methods.
  • These WMLES simulations demonstrate that state-of-the-art scale-resolving methods can provide accurate aerodynamic data for iced wings within turnaround times suitable for engineering applications.
  • Next steps include investigating different methods for incorporating ice shapes into the simulations using the hybrid body-fitted/immersed-boundary WMLES approach, and exploring alternative meshing.

Researchers

  • David A. Craig Penner, NASA Ames Research Center
  • Jeffrey A. Housman, NASA Ames Research Center

More Information