Devdiscourse
Development News Edition
Give Feedback
VisionRI - Urban Development

Faster development of clean fusion energy now possible with Artificial intelligence


Devdiscourse News Desk
Updated: 20-04-2019 22:06 IST
Faster development of clean fusion energy now possible with Artificial intelligence

Unlike traditional software, which carries out prescribed instructions, deep learning learns from its mistakes. Image Credit: Pixabay

Artificial intelligence could now speed the development of safe, clean and virtually limitless fusion energy for generating electricity, according to the findings reported in the current issue of Nature magazine.

A team of researchers from Princeton University and Harvard University are applying deep learning for the first time to forecast sudden disruptions that can halt fusion reactions and damage the doughnut-shaped tokamaks that house the reactions. Fusion is the reaction where two lighter atoms join to form a heavier one. It is the process that powers the sun and the stars. The deep learning code called the Fusion Recurrent Neural Network (FRNN) also opens possible pathways for controlling as well as predicting disruptions.

"Artificial intelligence is the most intriguing area of scientific growth right now, and to marry it to fusion science is very exciting," said William Tang, a principal research physicist at Princeton Plasma Physics Laboratory (PPPL). "We've accelerated the ability to predict with high accuracy the most dangerous challenge to clean fusion energy," he added.

"The ability of deep learning methods to learn from such complex data make them an ideal candidate for the task of disruption prediction," said collaborator Julian Kates-Harbeck, a physics graduate student at Harvard and a DOE-Office of Science Computational Science Graduate Fellow who was lead author of the Nature paper and chief architect of the code.

Unlike traditional software, which carries out prescribed instructions, deep learning learns from its mistakes. Accomplishing this seeming magic are neural networks that are "parameterized," or weighted by the program to shape the desired output. For any given input the nodes seek to produce a specified output, such as correct identification of a face or accurate forecasts of a disruption. Training kicks in when a node fails to achieve this task: the weights automatically adjust themselves for fresh data until the correct output is obtained.

While the researchers say that only live experimental operation can demonstrate the merits of any predictive method, their paper notes that the large archival databases used in the predictions, "cover a wide range of operational scenarios and thus provide significant evidence as to the relative strengths of the methods considered in this paper." The research opens a promising new chapter in the effort to bring unlimited energy to Earth.

(With inputs from Princeton University)


POST A COMMENT