In the framework of the Galaxy Serpent exercise a dataset of an unknown spent fuel material (code name “Terpsichore”) was compared against three datasets associated with three distinct reactors. The data of the known materials were compiled in a model national nuclear forensics library. The comparative evaluation aimed at (i) assessing whether the hypothetical seizure is or is not consistent with material contained in the library and (ii) identifying the reactor the unknown material may be associated with. Two different evaluation methods were applied. The first one is based on the isotope correlation technique and pointed at one of the three known reactors (code name “Siarnaq”) as potential origin of the unknown material. The second method uses multivariate analysis for clustering and classification of the data, identifying the same known reactor (Siarnaq) as the likely origin of the Terpsichore material. Thus, the evaluation resulted in a consistent answer provided by two independent methods. In addition to demonstrating the usefulness of having a national nuclear forensics library, the exercise showed that selected statistical techniques, such as principal component analysis or factor analysis may serve to optimize the content of a national nuclear forensics library by eliminating redundant (correlating) variables.
WALLENIUS Maria;
VARGA Zsolt;
MAYER Klaus;
2014-08-22
INSTITUTE OF NUCLEAR MATERIALS MANAGEMENT, INC.
JRC89433
0893-6188,
https://publications.jrc.ec.europa.eu/repository/handle/JRC89433,
| Name | Country | City | Type |
|---|
This document is only visible at the Commission level.
You are not authorized to publish or distribute it outside the European Commission.
This is a public document. You can share this publication.
Datasets
| ID | Title | Public URL |
|---|
Dataset collections
| ID | Acronym | Title | Public URL |
|---|
Scripts / source codes
| Description | Public URL |
|---|
Additional supporting files
| File name | Description | File type |
|---|