1. Home
  2. Project Highlights - 2023
  3. Lasers and Optical Science and Technology - 2023
  4. Predicting Damage Growth using Multimode Characterization and Machine Learning

Predicting Damage Growth using Multimode Characterization and Machine Learning

Jae Hyuck Yoo | 22-ERD-003

Executive Summary

We will develop a non-destructive, diagnostics-based strategy to predict the growth of laser-induced damage on optics by leveraging multi-modal characterization of damage sites and machine learning techniques. This effort will directly impact the long-term operational sustainability of fusion-class lasers, enabling them to operate more effectively, at higher energies.

Publications, Presentations, and Patents

Yoonsoo Rho, Christopher F. Miller, Robin E. Yancey, Ted A. Laurence, Christopher W. Carr, and Jae-Hyuck Yoo. "Wide-field probing of silica laser-induced damage precursors by photoluminescence photochemical quenching." Opt. Lett. 48, 3789-3792 (2023), LLNL-JRNL-847952.

Yoonsoo Rho, Christopher F. Miller, Robin E. Yancey, Ted A. Laurence, Christopher W. Carr, and Jae-Hyuck Yoo, "Temporally and spatially resolved photoluminescence of laser-induced damage sites of fused silica," (Abstract, SPIE Laser Damage Conference 2023). LLNL-ABS-848062.