Sign In

Communications of the ACM

ACM TechNews

Deep Learning Expands Study of Nuclear Waste Remediation

View as: Print Mobile App Share:
Nuclear Waste

Researchers at Lawrence Berkeley National Laboratory (Berkeley Lab), Pacific Northwest National Laboratory (PNNL), Brown University, and NVIDIA have achieved exaflop performance on the Summit supercomputer. The team applied this breakthrough to a deep learning model used to simulate subsurface flow in the study of nuclear waste remediation.

The researchers used synthetic data based on expert knowledge about the Hanford Site in Washington State. Using synthetic data enabled the researchers to create a virtual representation of the site that they could then manipulate as needed based on the parameters they were interested in measures.

Said Berkeley Lab researcher Prabhat, "We wanted to create an inexpensive surrogate for a very costly simulation, and what we were able to show here is that a physics-constrained GAN architecture can produce spatial fields consistent with our knowledge of physics."

From Berkeley Lab Computing Sciences
View Full Article


Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA


No entries found