Engineering staff fully exploit the massive computing power onsite and with the "big iron," unravel many wicked problems. Besides advanced M&S and data-analytic tools, entire fields of study, such as collaborative autonomy, are under incubation to enable groundbreaking solutions.
We develop physics and statistically-based models of systems, including extreme environments otherwise impossible to observe.
Our topics include wide-area motion analysis, pattern recognition, machine learning, computer-vision analysis, tomography, and radar and hyperspectral processing
LLNL is refining the role of the human guide in collaboration with an autonomous partner, as well as machine-to-machine collaboration.
We study optimization, machine learning, real-time processing, uncertainty quantification and data mining for static and dynamic structural, biological, and environmental systems.
Some of our current projects we are working on include:
Sierra Nevada is a series of explosive experiments in which engineers provide answers and solutions through virtual models of the equipment used. Engineers combine finite-element and finite-volume software codes to simulate the experiments, ensuring their safe and effective performance.
Insensitive High-Explosive Models
LLNL insensitive high-explosive numerical models calculate stress strain and damage from arbitrary thermomechanical loads, thereby enabling engineering assessments of plastic-bonded insensitive high explosives.
LiDO Design Optimization
Livermore design optimization (LiDO) structural-topology-optimization, high-performance-computing code is built on LLNL finite-element software, solvers, optimizers and visualization libraries. Fully parallel, it solves problems with one billion design parameters.
The Nuclear Survivability Program provides capabilities to determine whether a nuclear weapon can survive a nuclear intercept. A highly diverse team is breaking new ground daily using simulations and experiments to assay the viability of the U.S. nuclear deterrent.
Back to top