Case Study: Optimization of Wing Airflow Simulations
Overview
Computational fluid dynamics (CFD) is a key tool for simulating flight vehicles, including commercial aircraft, unmanned aerial vehicles (UAVs) and military aircraft. To be useful in the design process, these simulations must be efficient and reliable, with good estimates on the uncertainty in the computed results.
In collaboration with researchers at Virginia Tech, ANSYS Canada, and the US Air Force Research Laboratory, Dr. Ollivier-Gooch’s team is developing methods to compute more accurate, reliable results for complex aerodynamics problems an order of magnitude more efficiently than has previously been possible.
Funding
This project is funded by the
- US Air Force Office of Scientific Research
- ANSYS Canada
- NSERC
- US Air Force Research Laboratory
Solution
We are developing several new numerical tools on this project.
First, CFD solutions are inaccurate because of the finite resolution of the simulation. We are exploring different approaches for computing the solution error in our simulations more efficiently than simply repeating the simulation with finer spatial resolution.
Second, CFD simulations often fail to reach a steady state solution for numerical reasons. We are developing an approach for making small changes to the computational mesh so that a stable simulation is possible without the additional human cost of generating a new mesh (or more than one!).
Results
Much still remains to be done, but our results in both these areas have been quite encouraging to date. For example, our error estimation work can reduce error to match the error of more expensive high-order methods at a cost savings of about 20%; we expect to be able to reduce the cost of our method even further.
Before:
Close-up of error from typical CFD simulation for flow around the cross-section of a wing.
After:
Close-up of reduced error using our approach.
The Lab: The Advanced Numerical Simulation Laboratory
The mission of the Advanced Numerical Simulation Laboratory (ANSLab) is to improve computational techniques for the solution of partial differential equations, with a particular interest in application to transonic aerodynamics, where improved methodology for flow solution continues to bring higher-fidelity simulation into the engineering design loop. More specifically, members of the group work in three primary areas:
- high-order accuracy methods on unstructured meshes
- unstructured mesh generation and refinement, especially reducing the required amount of human input
- the interaction between mesh quality and solution accuracy.
The team working in the lab is interdisciplinary: knowledge of aerodynamics, computer science, and mathematics are all valuable to the lab.