Most Potent Tool in Fusion Science
Written by Tim Lash, Focus Fusion Society Contributor.
There are a pair of stories highlighting the importance of computer simulation for fusion science. Fusion research is expensive. Big reactors require large quantities of costly materials. Construction projects run for years or decades. Ramp up times necessitate huge investments before the science can even begin. Hence the usefulness of computer simulation. Careful modeling of reactor design or plasma behavior can yield insights that save time and money while inching us closer to clean fusion power. Therefore, computers are often the most valuable tool to the fusion scientist.
The first story describes simulation of plasma turbulence. Researchers used multi-scale simulations to study turbulent instabilities that cause plasma heat loss. The simulation added strong evidence that electron energy transport is indeed a multi-scale phenomenon. Being able to accurately predict electron energy transport is critical for predicting performance in future reactors such as ITER. These turbulence simulations used GYRO gyrokinetic plasma turbulence code at Lawrence Berkeley National Laboratory’s National Energy Research Scientific Computing Center (NERSC). The simulations used nearly 70 million hours of computing time on NERSC’s Edison system. Edison is a Cray XC30, with a peak performance of 2.57 petaflops/sec, 133,824 compute cores, 357 terabytes of memory, and 7.56 petabytes of disk.
A different simulation effort looked at the alpha particle products of fusion reactions. Alpha particles created by fusion carry a tremendous amount of kinetic energy. Under typical tokamak conditions these alpha particles can exhibit Alfvén eigenmodes. Alfvén eigenmodes are wave-like disturbances produced by the fusion reactions that ripple through the plasma. These waves can throw the alpha particles clear of the plasma. Escaping particles endanger the reactor chamber walls and efficient heating of the fuel. Physicists at the DOE’s Princeton Plasma Physics Laboratory (PPPL) produced an accurate model of the impact of these Alfvén waves on high-energy deuterium beams. They used simulation codes called NOVA and ORBIT to predict which Alfvén waves would be excited and their effects.
Focus Fusion Society posts have covered other simulation stories before. Of course, simulation assisted the work done at LPP Fusion as well. LPP computer simulations have confirmed the expected output beam energies as well as formation of the plasmoid.