Statistical modelling of neural networks in gamma-spectrometry applications - Archive ouverte HAL Access content directly
Conference Papers Year :

Statistical modelling of neural networks in gamma-spectrometry applications

(1, 2) , (3) , (3) , (1)
1
2
3

Abstract

Layered Neural Networds, which are a class of models based on neural computation, are applied to the measurement of uranium enriched, I.E. the isotope ratio 235U / (235U+236U+238U). The usual methods consider a limited number of gamma-ray and X-ray peaks, and require previously calibrated instrumentation for each sample. But, in practice, the source-detector ensemble geometry conditions are critically different, thus a means of improving the above conventional methods is to reduce the region of interest: this is possible by focusing on the KaX region where the three elementary components are present. The measurement of these components in mixtures leads to the desired ratio. Real data are used to study the performance of neural networks. Training is done with a Maximum Likelihood method. We show the encoding of data by Neural Networks is a promising method to measure uranium 235U and 238U quantities in infinitely thick samples.
Fichier principal
Vignette du fichier
article_VincentVigneron_ICRM95.pdf (401.86 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

cea-03956677 , version 1 (25-01-2023)

Identifiers

  • HAL Id : cea-03956677 , version 1

Cite

Vincent Vigneron, Jean Morel, Marie-Christine Lépy, Jean-Marc Martinez. Statistical modelling of neural networks in gamma-spectrometry applications. ICRM 95 - International Conference on Radionuclide Metrology and its applications, Commissariat à l'Energie Atomique; Bureau National de la Métrologie, May 1995, Paris, France. ⟨cea-03956677⟩
0 View
0 Download

Share

Gmail Facebook Twitter LinkedIn More