Next Generation High-Order Methods for Multi-Physics Multi-Scale Problems
Abstract
Computer simulation is an essential part of modern air-vehicle design. However, despite their numerous successes, current tools struggle to accurately predict the dynamics of highly turbulent flows with separation. Reliable and routine prediction of such flows, however, is a necessity for the design of next generation hypersonic vehicles and jet engines. This limitation is largely due to inadequacies with current generation numerical methods in terms of their ability to control dispersion and their suitability for massively parallel computers. High-order discontinuous spectral element methods are a promising candidate for resolving these issues, combining the superior accuracy of spectral schemes with the geometric flexibility of finite volume methods. However, in their current form they are unsuitable for many industrially-relevant problems, including those at supersonic and hypersonic speeds. The research objective of this proposal is to address these deficiencies through the development of new methods and technology which will enable the routine simulation of problems at high Reynolds numbers and those at high speeds. This will be accomplished by investigating the use of so-called meshless techniques combined with mixed-precision preconditioning schemes to accelerate the rate at which solutions are converged. The reliability of the schemes will be improved through a novel entropy filtering approach which has the potential to deliver superior robustness to existing approaches at a greatly reduced computational cost. The outcome of this work will be algorithms and software implementations thereof that can be used to efficiently solve a range of hitherto intractable flow problems. The resulting methods will be evaluated on a several highly strategic test cases including the NASA Rotor 37 case, the hypersonic BoLT geometry, and the NASA common research model high-lift configuration.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 29, 2024
- Source ID
- FA95502310232
Entities
People
- Freddie Witherden
Organizations
- Air Force Office of Scientific Research
- Texas Engineering Experiment Station
- United States Air Force