News
US Department of Energy Provides Almost $3M to use AI to Improve Inertial Fusion Energy Generation
Nuclear fusion has the potential to revolutionize power generation throughout the world as a source of Green energy. In inertial confinement fusion, a target filled with fuel (hydrogen isotopes) is compressed and heated to initiate nuclear fusion reactions. However, experiments at laser facilities such as OMEGA at Rochester’s Laboratory for Laser Energetics (LLE) generate less energy than simulations predict. Moreover, existing simulators are limited to predicting the outcome of the experiment, but do not have a way to generate a laser or target configuration that leads to the optimal amount of energy produced.
With PI Riccardo Betti, Co-PI Christopher Kanan, Associate Professor of Computer Science, will use artificial intelligence (AI) to improve the predictions of these simulations. Specifically, Kanan’s team will develop generative AI systems for inferring the optimal laser pulse shape and target configuration that yield the maximum amount of energy. His work builds upon recent works that have revolutionized the generation of images and video from text-based descriptions (e.g., OpenAI’s DALL-E and Midjourney), but repurposes this class of algorithm to solve inverse physics problems. The US Department of Energy is providing the team almost $3 million for this work, which is expected to be completed by 2026.
Read more here: https://www.hpe.com/us/en/newsroom/blog-post/2023/09/hewlett-packard-labs-receives-two-doe-fusion-initiative-awards-projects.html
https://www.photonics.com/Articles/Tower_Semiconductor_InnoLight_Pair_for_Silicon/a69306