Radionuclide classification based on gamma spectroscopy and artificial intelligence
DOI:
https://doi.org/10.15392/bjrs.v9i3.1653Palabras clave:
Espectroscopia gama, Inteligência artificial, Redes Neurais ArtificiaisResumen
Currently, in almost all segments of the production chain, automation is a requirement for productivity improvement. With respect to nuclear facilities, active online monitoring is one of best practices for nuclear security and safety maintenance, to prevent incidents that could compromise a particular installation. In this context, spectral signature monitoring automation can be explored, aiming at the rapid identification of adverse events, such as radiological accidents. The main objective of this work was an automated radionuclides classification technique establishment, using an Artificial Neural Networks (ANN) architecture. The methodology used consisted basically of simulating the geometry of an established experimental apparatus, using the MCNP5 code, obtaining the simulated gamma spectral signature for the studied nuclides. The simulated spectra were used to compose the ANN training and testing data set, while the experimental spectra were subjected to the artificial intelligence model classification, in order to allow the neural network quality assessment. The final developed architecture of ANN was correct to recognize the experimental spectra of 60Co, 137Cs and 152Eu. Therefore, the results were satisfactory and proved automation technique development viable.
Descargas
Referencias
X-5 Monte Carlo Team, MCNP - Version 5, Vol. I: Overview and Theory, LA-UR-03-1987 (2003).
ROSSUM, G. V., et al. Python Reference Manual. Python Software Foundation. 2020. Available at: < https://docs.python.org/3/download.html >. Last accessed: 31 Oct. 2020.
CHOLLET, F., et al. Keras. 2015. Available at: < https://github.com/fchollet/keras >. Last accessed: 31 Oct. 2020.
ABADI, M., et al. TensorFlow: Large-scale machine learning on heterogeneous systems. 2015. Available at: < https://www.tensorflow.org >. Last accessed: 31 Oct. 2020.
GOOGLE LLC. Colaboratory. Available at: <https://colab.research.google.com>. Last accessed: 31 Oct. 2020.
DAS R., CHAUDHURI S. On the Separability of Classes with the Cross-Entropy Loss Function. arXiv e-prints (Cornell University), 2019. Available at: <https://arxiv.org/pdf/1909.06930.pdf >. Last accessed: 02 Nov. 2020.
NUNES, W. V. Detecção de minas terrestres por radiação penetrante. Universidade Federal do Rio de Janeiro, 2005. 123p.
CURZIO, R. C., et al. Modelagem computacional da dispersão atmosférica aplicada a um reator modular de pequeno porte. Brazilian Journal of Radiation Sciences. 2020
PEREIRA, J.F. Explosão de bomba suja em local de grande evento público: uma metodologia para ações de emergência. CNEN/IRD, 2018. 124p.
HAYKIN, S. Neural Networks and Learning Machines, 3rd ed. New Jersey: Pearson Education Inc., 2009.
KINGMA, D. P., BA, J. L. Adam: A Method for Stochastic Optimization. In: International Conference on Learning Representations, 2015, San Diego. Annals… San Diego, 2015. Available at: < https://arxiv.org/pdf/1412.6980.pdf>. Last accessed: 02 Nov. 2020.
TSOULFANIDIS, N., LANDSBERGER, S. Measurement & Detection of Radiation, 4rd ed. Boca Raton: Taylor & Francis Group, 2015.
Publicado
Número
Sección
Licencia
Derechos de autor 2021 Brazilian Journal of Radiation Sciences (BJRS)
Esta obra está bajo una licencia internacional Creative Commons Atribución 4.0.
Licencia: los artículos de BJRS tienen una licencia internacional Creative Commons Attribution 4.0, que permite el uso, el intercambio, la adaptación, la distribución y la reproducción en cualquier medio o formato, siempre que se otorgue el crédito correspondiente al autor o autores originales y a la fuente, proporcione un enlace a la licencia Creative Commons e indique si se realizaron cambios. Las imágenes u otros materiales de terceros en el artículo están incluidos en la licencia Creative Commons del artículo, a menos que se indique lo contrario en una línea de crédito al material. Si el material no está incluido en la licencia Creative Commons del artículo y su uso previsto no está permitido por la regulación legal o excede el uso permitido, el autor deberá obtener el permiso directamente del titular de los derechos de autor. Para ver una copia de esta licencia, visite http://creativecommons.org/licenses/by/4.0/