The idea of using powerful simulations to test how things explode sounds incredibly enjoyable — until you’re faced with the painstaking task of creating line after line of meticulous code. At that point, blowing up the simulator probably sounds more enticing. But a joint project between simulation software engineering company ANSYS and the Army aims to wrangle complex government code into an easier-to-use format.
The Army Armament Research, Development and Engineering Center collaborated with ANSYS to create an interface that wraps around complex codes used for government simulations, many of which are used for defense research.
Why bother wrapping the code in a pretty package? The government focuses on power and capability for its sims — but not necessarily usability. Assembling correct code to run the simulations can take days or weeks and scare off less experienced engineers. ANSYS software works toward a streamlined, push-button approach to generate the code necessary to run the simulation, but in a faster and more user-friendly way.
The creative partnership has spent the last year making ADAPT, the ARDEC/ANSYS Developed Analysis Preprocessing Tool, and applying it to government code from Lawrence Livermore National Laboratory. The physics code models all manner of things that defense research laboratories are interested in: detonations, how objects fracture upon impact, hydrodynamics, heat transfer, magnetic forces, and structural interactions.
If you want to model things that go boom or the things that they hit, this is the code that does it. For the military, research on these topics can change the body armor, weapons, and even tactics that are used in war.
Developers hope that by simplifying and speeding up the process, researchers will be able to increase productivity, spread tasks out to more junior members, and use simulations at more locations. While the LLNL code is the only completed project that ADAPT has been applied to, developers are also working with Sandia National Laboratories and say the technology could be applied to any powerful but unwieldy code.
“It doesn’t matter if it’s the Army or the Navy or the Air Force,” said Anthony Dawson, senior technical services engineer at ANSYS and a former Defense Department research scientist. “They all have codes that are candidates for this; they all have research sites with users that are using the codes that are candidates for this. It’s a very broad spectrum.”
Dawson said that other types of code could include simulations covering general structural mechanics, explicit dynamics, or computational fluid dynamics. The terms are complicated, but a practical example is easy to find. Such code could be used to model air resistance as it flows over the wing of a fighter jet as researchers search to increase the power and speed of the next generation of weapons.
If applied to such code, the ADAPT tool presents whoever is setting up the simulation with a more intuitive user interface that ANSYS has developed. It is customized to easily set up the code for that particular simulation. Set up the parameters for the sim, and the software will spit out an input deck that can be fed into the original simulation.
“You can execute your simulation with government code in the same way you would have if you wrote that input deck, but you didn’t write it. We wrote it for you,” Dawson said.
ADAPT is also designed to be particularly adaptable. If a certain function isn’t preprogrammed into the software, developers can write it in themselves. It’s like assembling a salad: you can plop in the basic ingredients you bought from the store, but add in a homemade vinaigrette if you decide the supermarket version just doesn’t cut it. Ultimately, it’s a lot faster than growing the lettuce and tomatoes yourself.
ARDEC is the flagship user and primary customer at this point, though anyone with ANSYS software and the Livermore code could use the ADAPT technology. According to Dawson, feedback has been positive, with ARDEC reporting an increase in productivity and other unnamed DoD sites taking on the tool as early adopters.
Dawson noted that usability issues have plagued government code for some time, but not everyone will embrace their solution. One potential problem will ring true with those who have pushed simulation technology in the past.
“The only downside, I would say, is some people are set in their ways,” he said.