Enabling high-fidelity spectroscopic analysis of plutonium with machine learning
Document Type
Conference Proceeding
Publication Date
7-15-2022
Abstract
Machine learning methods are constructed to perform analysis of plutonium surrogate material. Decision tree based methods yield predictive models for quantifying gallium from optical emission spectra with sensitivities as low as 0.006 wt%.
Source Publication
Optical Sensors and Sensing Congress 2022 (AIS, LACSEA, Sensors, ES)
Recommended Citation
Rao, A. P., Jenkins, P. R., & Patnaik, A. K. (2022). Enabling high-fidelity spectroscopic analysis of plutonium with machine learning. Optical Sensors and Sensing Congress 2022 (AIS, LACSEA, Sensors, ES), LF1C.1. https://doi.org/10.1364/LACSEA.2022.LF1C.1
Comments
The full text of this paper is available via subscription or purchase at Optica, through the DOI link below.
Co-author Ashwin Rao was completing his AFIT PhD program at the time of this publication. (AFIT-ENP-DS-22-S-048, September 2022)