Scientists have just taken a giant step towards expanding the scope of nuclear fusion

A team of researchers at MIT believe they may have lowered one of the major barriers to achieving large-scale nuclear fusion, bringing us one step closer to making an abundant form of energy a reality.

By harnessing the same processes that power stars, we will have access to a clean, safe, and practically unlimited energy source. Scientists have built reactors to try to tame nuclear fusion, and one of the most explored is the tokamak. The tokamak, essentially a donut-shaped tube that uses powerful magnets to confine the plasma needed to power nuclear fusion reactions, has shown great potential. But to fully realize this, scientists must first explore the potential dangers this energy carries, including how to slow down the fusion reaction once it occurs.

This is where New search It comes: Using a combination of physics and machine learning, researchers predicted how the plasma inside a tokamak reactor would behave under a set of initial conditions, something that has long puzzled researchers (it’s hard to look inside a fusion reactor mid-run, after all). The paper was published Monday in the journal Nature Communications.

“For fusion to become a useful source of energy, it must be reliable,” said Allen Wang, lead author of the study and a graduate student at MIT. MIT News. “To be trustworthy, we have to get good at managing our plasma.”

With great power comes great risks

When a tokamak reactor is operating at full capacity, the stream of plasma inside it can rotate at speeds of about 62 miles (100 kilometers) per second and at temperatures of up to 180 million degrees Fahrenheit (100 million degrees Celsius). This is hotter than the core of the sun.

If the reactor must be shut down for any reason, operators begin a process to “tamp down” the plasma stream, slowly de-energizing it. But the process is difficult, and the plasma can cause “scratches and scars on the interior of the tokamak, minor damage that still requires significant time and resources to repair,” the researchers explain.

“Uncontrolled plasma terminations, even during reduction, can generate intense heat fluxes that damage internal walls,” Wang explained. “Often, especially with high-performance plasmas, rollback can actually push the plasma closer to some instability limit. So, it’s a delicate balance.”

In fact, any error in the operation of nuclear fusion reactors could be costly. In an ideal world, researchers would be able to conduct tests in tokamak operation, but since fusion is still inefficient, operating one of these reactors is incredibly expensive, and most facilities will only run it a few times a year.

Looking to the wisdom of physics

For their model, the team found A A delightfully clever way To overcome the limitations of data collection, they simply returned to the basic rules of physics. They connected their model’s neural network to another model that describes plasma dynamics, and then trained the model on data from TCV, a small experimental fusion device in Switzerland. The data set included information about variations in plasma starting temperature and energy levels, as well as during and at the end of each experimental run.

From there, the team used an algorithm to generate “trajectories” that showed reactor operators how the plasma was likely to behave as the reaction progressed. When they applied the algorithm to actual TCV operations, they found that following the model’s “path” instructions was perfectly capable of guiding operators to safely unload the device.

“We’ve done it several times,” Wang said. “And we did things much better across the board. So, we had statistical confidence that we made things better.”

“We are trying to address the scientific issues to make fusion routinely useful,” he added. “What we’ve done here is the beginning of a long journey. But I think we’ve made some good progress.”