Turbulence model could enhance rotorcraft, munitions performance

The simulations that designers currently use require extensive data processing on supercomputers and capture only a portion of vortex collision events -- which can cause significant performance degradation, from loss of lift on a rotor to complete loss of control of a munition. A new turbulence model could change that.

The Army Research Office, an element of the U.S. Army Combat Capabilities Development Command, now known as DEVCOM, Army Research Laboratory, funded researchers at Purdue University to advance a turbulence model known as the Coherent-vorticity-Preserving Large-Eddy Simulation, known as CvP LES. Published in the Journal of Fluid Mechanics, the new methodology simulates the entire process of a vortex collision event up to 100 times faster than current state-of-the-art simulation techniques.

"The thing that's really clever about Purdue's approach is that it uses information about the flow physics to decide the best tactic for computing the flow physics," said Dr. Matthew Munson, Program Manager for Fluid Dynamics at ARO. "There is enormous potential for this to have a real impact on the design of vehicle platforms and weapons systems that will allow our Soldiers to successfully accomplish their missions."

The fluid dynamics of aircraft turbulence are complex, and simulating them accurately in the computer is nearly impossible. Prof. Carlo Scalo has taken a leap forward in this process, by modeling the collision of vortices in two ways: once with direct numerical simulation, and once with large-eddy simulation. This model can now be used by engineers to design better aircraft, without having to wait months for supercomputer calculations. Carlo Scalo's Compressible Flow and Acoustics Lab: https://engineering.purdue.edu/~scalo/ Mechanical Engineering: https://purdue.edu/ME

The model can be used to simulate vortices over any length of time to best resemble what happens around an aircraft. For instance, as a rotor blade moves through the air, it generates a complex system of vortices that are encountered by the next blade passage. The interaction between the blade and the vortices can lead to vibration, noise, and degraded aerodynamic performance. Understanding these interactions is the first step to modifying designs to reduce their impact on the vehicle's capabilities.

In this study, researchers simulated the collision events of two vortex tubes called trefoil knotted vortices. This interaction shares many common features to the vortices often present in Army applications. Simulating the evolution of the collision requires extremely fine resolution, substantially increasing the computational cost.

The methodology relies on clever techniques that balance cost and accuracy. It is capable of rapidly detecting regions of the flow characterized by fine turbulent scales and then determining, on-the-fly, the appropriate numerical scheme and turbulence model to apply locally. This also allows computational power to be applied only where most needed, achieving a solution with the highest possible fidelity for a given budgeted amount of computational resources.

"When vortices collide, there's a clash that creates a lot of turbulence," said Carlo Scalo, a Purdue associate professor of mechanical engineering with a courtesy appointment in aeronautics and astronautics. "It's very hard computationally to simulate because you have an intense localized event that happens between two structures that look pretty innocent and uneventful until they collide."

Using the Brown supercomputer at Purdue University for mid-size computations and Department of Defense facilities for large-scale computations, the team simulated an entire collision event, fully simulating the thousands of events that take place when these vortices collide.

The team is now working with the Department of Defense to apply the model to large-scale test cases pertaining to Army vehicle and weapons systems.

"If you're able to accurately simulate the thousands of events in flow like those coming from a helicopter blade, you could engineer much more complex systems," Scalo said.

The Rosen Center for Advanced Computing at Purdue and the U.S. Air Force Research Laboratory Department of Defense Supercomputing Resource Center provided additional support for this research.