Tokamaks are machines that are supposed to hold and harness the ability of the sun. These fusion machines use powerful magnets to contain a plasma hotter than the sun’s core and push the plasma’s atoms to fuse and release energy. If tokamaks can operate safely and efficiently, the machines could someday provide clean and limitless fusion energy.
Today, there are a variety of experimental tokamaks in operation around the globe, with more underway. Most are small-scale research machines built to research how the devices can spin up plasma and harness its energy. One in every of the challenges that tokamaks face is the best way to safely and reliably turn off a plasma current that’s circulating at speeds of as much as 100 kilometers per second, at temperatures of over 100 million degrees Celsius.
Such “rampdowns” are essential when a plasma becomes unstable. To stop the plasma from further disrupting and potentially damaging the device’s interior, operators ramp down the plasma current. But occasionally the rampdown itself can destabilize the plasma. In some machines, rampdowns have caused scrapes and scarring to the tokamak’s interior — minor damage that also requires considerable time and resources to repair.
Now, scientists at MIT have developed a technique to predict how plasma in a tokamak will behave during a rampdown. The team combined machine-learning tools with a physics-based model of plasma dynamics to simulate a plasma’s behavior and any instabilities which will arise because the plasma is ramped down and turned off. The researchers trained and tested the brand new model on plasma data from an experimental tokamak in Switzerland. They found the tactic quickly learned how plasma would evolve because it was tuned down in alternative ways. What’s more, the tactic achieved a high level of accuracy using a comparatively small amount of information. This training efficiency is promising, on condition that each experimental run of a tokamak is pricey and quality data is restricted because of this.
The brand new model, which the team highlights this week in an open-access paper, could improve the protection and reliability of future fusion power plants.
“For fusion to be a useful energy source it’s going to should be reliable,” says lead writer Allen Wang, a graduate student in aeronautics and astronautics and a member of the Disruption Group at MIT’s Plasma Science and Fusion Center (PSFC). “To be reliable, we want to get good at managing our plasmas.”
The study’s MIT co-authors include PSFC Principal Research Scientist and Disruptions Group leader Cristina Rea, and members of the Laboratory for Information and Decision Systems (LIDS) Oswin So, Charles Dawson, and Professor Chuchu Fan, together with Mark (Dan) Boyer of Commonwealth Fusion Systems and collaborators from the Swiss Plasma Center in Switzerland.
“A fragile balance”
Tokamaks are experimental fusion devices that were first inbuilt the Soviet Union within the Fifties. The device gets its name from a Russian acronym that translates to a “toroidal chamber with magnetic coils.” Just as its name describes, a tokamak is toroidal, or donut-shaped, and uses powerful magnets to contain and spin up a gas to temperatures and energies high enough that atoms within the resulting plasma can fuse and release energy.
Today, tokamak experiments are relatively low-energy in scale, with few approaching the scale and output needed to generate secure, reliable, usable energy. Disruptions in experimental, low-energy tokamaks are generally not a problem. But as fusion machines scale as much as grid-scale dimensions, controlling much higher-energy plasmas in any respect phases might be paramount to maintaining a machine’s secure and efficient operation.
“Uncontrolled plasma terminations, even during rampdown, can generate intense heat fluxes damaging the inner partitions,” Wang notes. “Very often, especially with the high-performance plasmas, rampdowns actually can push the plasma closer to some instability limits. So, it’s a fragile balance. And there’s plenty of focus now on the best way to manage instabilities in order that we will routinely and reliably take these plasmas and safely power them down. And there are relatively few studies done on the best way to do this well.”
Bringing down the heart beat
Wang and his colleagues developed a model to predict how a plasma will behave during tokamak rampdown. While they may have simply applied machine-learning tools resembling a neural network to learn signs of instabilities in plasma data, “you would want an ungodly amount of information” for such tools to discern the very subtle and ephemeral changes in extremely high-temperature, high-energy plasmas, Wang says.
As a substitute, the researchers paired a neural network with an existing model that simulates plasma dynamics in accordance with the basic rules of physics. With this mix of machine learning and a physics-based plasma simulation, the team found that only a pair hundred pulses at low performance, and a small handful of pulses at high performance, were sufficient to coach and validate the brand new model.
The information they used for the brand new study got here from the TCV, the Swiss “variable configuration tokamak” operated by the Swiss Plasma Center at EPFL (the Swiss Federal Institute of Technology Lausanne). The TCV is a small experimental fusion experimental device that’s used for research purposes, often as test bed for next-generation device solutions. Wang used the info from several hundred TCV plasma pulses that included properties of the plasma resembling its temperature and energies during each pulse’s ramp-up, run, and ramp-down. He trained the brand new model on this data, then tested it and located it was capable of accurately predict the plasma’s evolution given the initial conditions of a selected tokamak run.
The researchers also developed an algorithm to translate the model’s predictions into practical “trajectories,” or plasma-managing instructions that a tokamak controller can routinely perform to as an illustration adjust the magnets or temperature maintain the plasma’s stability. They implemented the algorithm on several TCV runs and located that it produced trajectories that safely ramped down a plasma pulse, in some cases faster and without disruptions in comparison with runs without the brand new method.
“Sooner or later the plasma will at all times go away, but we call it a disruption when the plasma goes away at high energy. Here, we ramped the energy right down to nothing,” Wang notes. “We did it a variety of times. And we did things significantly better across the board. So, we had statistical confidence that we made things higher.”
The work was supported partially by Commonwealth Fusion Systems (CFS), an MIT spinout that intends to construct the world’s first compact, grid-scale fusion power plant. The corporate is developing a demo tokamak, SPARC, designed to supply net-energy plasma, meaning that it should generate more energy than it takes to heat up the plasma. Wang and his colleagues are working with CFS on ways in which the brand new prediction model and tools like it may possibly higher predict plasma behavior and forestall costly disruptions to enable secure and reliable fusion power.
“We’re attempting to tackle the science inquiries to make fusion routinely useful,” Wang says. “What we’ve done here is the beginning of what remains to be a protracted journey. But I feel we’ve made some nice progress.”
Additional support for the research got here from the framework of the EUROfusion Consortium, via the Euratom Research and Training Program and funded by the Swiss State Secretariat for Education, Research, and Innovation.
