Nuclear fusion for clean and unlimited energy, AI reveals how to do it

It is possible to have clean and unlimited energy.
This is what artificial intelligence says, or rather the scientists who have questioned AI to overcome the energy challenge of the future and created a specific algorithm.
Researchers at Princeton University in the United States have built a model with artificial intelligence to predict and prevent plasma instability during nuclear fusion reactions.
This could lead to clean energy production, nuclear energy, without the risks associated with its production.
For many, nuclear fusion can represent the solution to the climate crisis and even if it is not an opinion that all experts agree on – because climate change is a current emergency characterized by more short-term challenges and targets than large-scale development scale of a new nuclear technology – the potential for the spread of dedicated power plants opens up the possibility of decarbonisation.
read also Nuclear energy: is fusion really dangerous? Fusion energy: the critical points Researchers from Princeton University in New Jersey spoke in the scientific journal Nature to talk about the progress achieved with artificial intelligence in the field of nuclear energy.
They have found a way to use AI to prevent plasma instability during fusion reactions.
What is it about? The process mimics the same natural reactions that occur inside the Sun, a process that is difficult to exploit inside a power plant.
In the Sun, nuclear fusion occurs when two or more atomic nuclei combine to form a larger, heavier atom.
The atom resulting from the fusion has a lower mass than the sum of the starting atoms: this difference in mass, in the process, is transformed into energy.
Scientists have long been trying to replicate how a star works in the laboratory.
A difficult operation, because it is unstable.
Among the obstacles: the increase in energy, increasingly high-performance materials and a reactor free from impurities.
read also Nuclear energy generated in the EU is at its lowest level since 1990 Fusion energy: the discovery explained This is where artificial intelligence comes into play.
Researchers have developed an AI model that can theoretically solve some of the problems.
The main discovery sees the algorithm prevent the plasma from becoming unstable and escaping the magnetic fields that hold it inside the reactors.
The artificial intelligence developed by the US university appears to be able to predict instabilities, such as a sudden change in magnetic current and its consequences.
The prediction occurs on small numbers, in fact the AI is capable of recognizing plasma instability 300 milliseconds before they occur.
Being able to predict instabilities in advance may make performing these reactions easier than current approaches, which are more passive, said SanKyeun Kim, a co-author of the study.
The success of the machine learning model comes from the fact that it was trained on real data from previous fusion experiments.
Over time the algorithm will learn the optimal path to reach the goal.
The potential to gain control is enormous, commented Federico Felici, a physicist at the Swiss Federal Institute of Technology.