Team pounds pavement
with concrete calculations

(page 3 of 3)

In this immersive visualization environment, Satterfield and George have demonstrated a volume of flowing concrete, with moving rocks surrounded by an invisible fluid, and mortar. Most simulations of complex mixtures use uniformly shaped particles such as spheres or ellipses. That's impractical in concrete models because rock shapes can profoundly affect flow properties. For example, concrete that uses smooth riverbed rocks flows more easily than concrete that incorporates rough-hewn quarry rock. The group's simulation now includes realistic models with varying sizes and rough shapes that John Hagedorn and Garboczi of NIST developed from images of real rocks captured with X-ray microtomography.

But simulating the interactions of irregularly shaped particles adds a level of computational bookkeeping to track the points and shapes of the rocks, George says. With a spherical particle, for example, the researchers just needed to know the position of its center and its radius. The extra details of realistic rock shapes have increased the computational size of these simulations 10-fold. The group relies on visualization to troubleshoot the initial calculations and see that rocks aren't colliding or moving through each other in space.

The immersive environment and its associated tools let the researchers look at the system as a whole and examine interactions layer by layer, frame by frame. They can label and graph the stresses on each rock and series of rocks, then color-code them accordingly, giving greater stresses a darker color.

Using visualizations, George says, the NIST team has observed that groupings of larger rocks tend to be under the most stress, and the physical mechanisms don't match what's been reported. Because groups of rocks appear to carry the stress, their arrangement is important. The compression and extension of materials in the flow also may contribute to the overall yield stress in the system.

"Visualization suggests a new way to analyze the data," says Martys, the NIST physicist. "It gives you focus."

« Previous       1   |   2   |   3   |   Print