AEM faculty spotlight:
As the accuracy and cost-effectiveness of computer-based simulations continues to increase, so too does its real-world impact. Government and industry are relying more on these types of methods and simulations for mission planning and related research, and AEM is at the forefront of these methods. AEM Assistant Professor Thomas Schwartzentruber hopes to bring a new capability and perspective to the mix.
Below, Schwartzentruber discusses his research and its potential applications.
Thomas E. Schwartzentruber
On what does your research focus?
My research utilizes particle simulation methods to model non-equilibrium gas flows that are found at very high altitudes and at very small scales. The key issue with these flows is that fewer gas particles are present and more importantly, they collide less frequently. When this happens, the gas no longer behaves like a continuous fluid and conventional Computational Fluid Dynamic (CFD) methods become inaccurate.
Could you discuss the relevance of this type of research?
Well, the current applications of this research relate to high altitude aerothermodynamics – such as the conditions experienced by a reentry vehicle entering the atmosphere from space. During the high-altitude portion of the trajectory, CFD methods do not accurately predict the heat transfer, drag, and aerodynamic moments on the vehicle. New missions are also attempting high-altitude “aerocapture” maneuvers and large decelerator mechanisms for landing large payloads on Mars. Basically, there is interest in using the upper atmosphere to slow down a spacecraft instead of carrying fuel to do it and possibly reducing the peak heating as well.
What projects are you working on?
Primarily I am focusing on the development of multi-scale methods. This combines conventional CFD methods with particle methods. I developed such a hybrid method for my PhD dissertation and it’s certainly an ongoing research interest. A reentry capsule is a great example of where a multi-scale method is necessary. What happens is that the forebody of the capsule has extremely high densities of gas particles but the wake, directly behind the capsule, contains very few particles. So you have continuum conditions at the front and non-equilibrium conditions in the wake. Currently, it’s very difficult even on supercomputers to carry out a full reentry particle-simulation because of the extreme densities in the forebody region. The idea of my research is to use CFD for the forebody –where it’s perfectly valid— and only utilize a particle method in a local way when and where it is appropriate. The goal is to accurately simulate all regions but in a more computationally efficient manner.
Tell me a bit about particle methods.
I guess the main difference compared to a continuum method is that instead of solving a set of partial differential equations, a particle method moves and collides simulation particles which carry and exchange mass, momentum, and energy with each other and the vehicle surface. In the continuum limit, both methods produce the same result, however, a particle-based method models the actual physics of the flow more closely. When you think about it, many of the interesting phenomena in aerospace flows, such as chemical reactions and surface heating are the result of high-energy collisions between gas molecules and surface molecules. In terms of what I am currently doing, another ongoing project is the continual development of the particle method I use, the Direct Simulation Monte Carlo (DSMC) method. DSMC is a type of particle method that works very well for high-speed, high-altitude flows and has been used by NASA, the Air Force, and industry for some time. However, DSMC is not as mature as CFD and there’s a lot of development that still needs attention. Things like local time-stepping, adaptive mesh refinement, and load-balancing on large parallel clusters are vital for an efficient DSMC code.
What do you bring to the faculty?
The bottom-line is that I bring the capability of particle simulation methods to the department. This will hopefully strengthen the National Center for Hypersonics Research so that AEM can provide support for entire spacecraft missions across all flight regimes. At the micro-scale, there is also great faculty expertise in AEM on atomistic materials modeling.
How would your research apply to faculty work in Solid Mechanics?
In particle simulations, where the gas is modeled as a group of atoms and molecules, you’ve now reached the same scale used by some of the Solids faculty who analyze and model materials at the atomistic level. This collaboration would really be great, and may allow the possibility of coupling fluid and solid mechanics methods. This could be useful in hypersonics for modeling ablation and surface catalysis and also very useful for microflows and biomedical flows. The interaction, for instance, between a MEMS device and its environment is very important.
Last Modified: Monday, 29-Aug-2016 10:36:14 CDT -- this is in International Standard Date and Time Notation