Artificial intelligence improves control of powerful plasma accelerators

Artificial intelligence improves control of powerful plasma accelerators

The gas cell used as a source of plasma. The laser comes from the right of these images through the metal cone and into the small cube, which is filled with gas. The laser ionizes the gas and turns it into plasma and creates the accelerator. Credit: Rob Shalloo

Researchers have used artificial intelligence to control the beams of the next generation of smaller and cheaper accelerators for research, medical and industrial applications.

Experiments led by researchers from Imperial College London, using the Council of Science and Technology Facilities Central Laser Facility (CLF), showed that an algorithm could fine-tune the complex parameters involved in controlling the next generation of particle-based accelerators. In plasma.

The algorithm was able to optimize the throttle much faster than a human operator and could even outperform experiments on similar laser systems.

These accelerators focus the energy of the world’s most powerful lasers to a point the size of a skin cell, producing electrons and X-rays with equipment a fraction of the size of conventional accelerators.

Electrons and X-rays can be used for scientific research, such as probing the atomic structure of materials; in industrial applications, such as the production of consumer electronics and vulcanized rubber for automobile tires; and it could also be used in medical applications such as cancer treatments and medical imaging.

Several facilities using these new accelerators are in various stages of planning and construction around the world, including CLF’s Extreme Photonics Applications Center (EPAC) in the UK, and the new discovery could help them perform at their best. way in the future. The results are published today in Communications from nature.

Artificial intelligence improves control of powerful plasma accelerators

Electrons are ejected from the plasma accelerator at almost the speed of light, before passing through a magnetic field that separates the particles by its energy. They are then shot at a fluorescent screen, shown here. Credit: Rob Shalloo

First author Dr. Rob Shalloo, who completed the work at Imperial and is now at the DESY Acceleration Center, said: “The techniques we have developed will be critical to getting the most out of a new generation of advanced accelerator facilities. Plasma under construction in the UK and around the world.

“Plasma accelerator technology provides exceptionally short X-ray and electron bursts, which are already finding uses in many areas of scientific study. With our developments, we hope to expand the accessibility of these compact accelerators, allowing scientists in other disciplines to Those who want to use these machines for applications, to benefit from the technology without being an expert in plasma accelerators. “

The team worked with wakefield laser accelerators. These combine the world’s most powerful lasers with a source of plasma (ionized gas) to create concentrated beams of electrons and X-rays. Traditional accelerators need hundreds of meters to miles to accelerate electrons, but wakefield accelerators can handle the same acceleration within the millimeter space, dramatically reducing equipment size and cost.

However, because wakefield accelerators operate in the extreme conditions created when lasers are combined with plasma, they can be difficult to control and optimize for the best performance. In wakefield acceleration, an ultrashort laser pulse is propelled into the plasma, creating a wave that is used to accelerate the electrons. Both laser and plasma have several parameters that can be modified to control the interaction, such as the shape and intensity of the laser pulse, or the density and length of the plasma.

While a human operator can modify these parameters, it is difficult to know how to optimize so many parameters at once. Instead, the team turned to artificial intelligence and created a machine learning algorithm to optimize throttle performance.

Artificial intelligence improves control of powerful plasma accelerators

This photo shows the exterior of the vacuum chamber, which is completely surrounded by painted lead bricks. The lead is for radiation protection and the metal frame allows the lead walls to roll in and out to allow access to the chamber. They are painted because lead is highly toxic, so painting them prevents them from generating harmful lead dust. Credit: Rob Shalloo

The algorithm configured six parameters that controlled the laser and plasma, fired the laser, analyzed the data, and reset the parameters, performing this loop many times in a row until the optimal parameter settings were reached.

Lead Investigator Dr Matthew Streeter, who completed the work at Imperial and is now at Queen’s University Belfast, said: “Our work resulted in an autonomous plasma accelerator, the first of its kind. As well as allowing us to optimize efficiently throttle, it also simplifies its operation and allows us to put more effort into exploring the fundamental physics behind these extreme machines. “

The team demonstrated their technique using the Gemini laser system at the CLF and have already begun using it in further experiments to probe the atomic structure of materials under extreme conditions and in the study of quantum and antimatter physics.

The data collected during the optimization process also provided new insight into the dynamics of laser-plasma interaction within the throttle, which could inform future designs to further improve throttle performance.

World record: plasma accelerator works 24 hours a day

More information:
RJ Shalloo et al. Automation and control of laser wakefield accelerators using Bayesian optimization, Communications from nature (2020). DOI: 10.1038 / s41467-020-20245-6

Provided by Imperial College London

Citation: Artificial Intelligence Improves Control of Powerful Plasma Accelerators (2020, December 11) Retrieved December 13, 2020 from

This document is subject to copyright. Apart from any fair dealing for private study or research purposes, no part may be reproduced without written permission. The content is provided for informational purposes only.