[1B13-17] Analytical Technique

[1B17] Development of Quick and Remote Analysis for Severe Accident Reactor-8

(5)Analysis of Laser-Induced Breakdown Spectra Using Machine Learning

*Katsuaki Akaoka¹, Masaki Ohba¹, Takahiro Karino¹, Hironori Ohba¹, Ikuo Wakaida¹ (1. JAEA)

Keywords:Fuel debris, machine learning, LIBS, laser, Laser-Induced Breakdown Spectroscopy, quantitative analysis, uranium, plutonium

In response to on-site analysis of fuel debris and other materials inside the reactor, which resulted from the accident at the Fukushima Daiichi Nuclear Power Station (FDNPS), we are evaluating and examining the application of Laser-Induced Breakdown Spectroscopy (LIBS) for quantitative analysis. Since the use of calibration curves created in laboratories may not be appropriate in this case, we are exploring quantitative analysis using machine learning techniques. In this report, we present the results of our attempt to perform quantitative analysis using machine learning based on the spectra of U/Pu mixed samples.