At the Princeton Plasma Physics Laboratory a team led by C.S. Chang, leader of the International XGC Program, is exploring the feasibility of realizing the seemingly unattainable dream of harnessing fusion to provide the whole of humanity with limitless, clean, sustainable electricity “for millions of years”.
In the video below, Chang explains that fusion is what fuels the sun and is the form of energy that scientists want to replicate.
Fusion involves heating nuclei of light atoms to create a plasma consisting of ionized particles. Once these are heated to millions of degrees they are forced to collide and join together to form a heavier nucleus which releases neutrons, giving off huge amounts of energy in the process.
The problem is that plasma is unstable and too hot at the edges for known materials to contain it. To find a solution the International Thermonuclear Experimental Reactor (ITER) megaproject was established to build a tokamak fusion reactor, a donut-shaped device in the south of France, which uses a powerful magnetic field to try to confine hot plasma to the center.
Chang’s aim is to analyze the problem through simulation models rather than trial and error.
“We are trying to understand plasma behavior, the most complicated physics area in fusion reactors, and we need the world’s biggest computer to do it,” he says.
That supercomputer is Summit. Powered by NVIDIA V100 GPUs, Summit was funded by the US Department of Energy and installed at the Oak Ridge National Laboratory in Tennessee last year and is currently the world’s most powerful supercomputer.
Chang says its sheer size and use of Volta GPUs enable his team to run their XGC code almost four times faster than on Titan.
The ultimate aim is to build a whole-device predictive model that within the next few years will help design and commission a fusion reactor that produces more energy than it consumes, creates no greenhouse gases, and produces no waste.
Be the first to comment