EUVIMEDEUVIMED entdecken
Crossref

Fast Generation of Gamma-Ray Logs Based on MCNP Simulations for Uranium Mining

Arthur Pellet-Rostaing, Bertrand Pérot, Thomas Marchais, Nadia Pérot, Hervé Toubon, Youcef Bensedik, Sebastien Hocquet, Christophe Pouet, Mitra Fouladirad

IEEE Transactions on Nuclear Science · 2026-06 · 73 (6) · 2224-2231

DOI: 10.1109/tns.2026.3672048 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Nuclear Science, 2026-06. Verfasst von Arthur Pellet-Rostaing und 8 weiteren Autor:innen. Die Arbeit nennt 14 Literaturangaben. Gefördert von: Research and Development Collaboration between Commissariat à l'Energie Atomique et aux énergies alternatives (CEA) and Orano Mining.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Orano Mining is evaluating the potential of a CeBr3 spectrometric gamma-ray logging probe for estimating uranium concentration in roll-front deposits, where decay chain disequilibrium disrupts its relation with total gamma count rate. The Nuclear Measurement Laboratory of CEA IRESNE is developing predictive algorithms that exploit the full spectral shape. Unfortunately, logged wells are currently too scarce to train them. Creating a representative database of simulated gamma-ray logs, using the Monte Carlo N-particle (MCNP) transport code, is therefore required. Conventional analog simulations based on the MCNP pulse-height (F8) tally provide accurate estimates, but are too computationally expensive. We propose two alternative methods for fast generation of gamma-ray logs. The first is a two-step scheme that decouples flux transport to the probe, estimated with the MCNP point detector (F5) tally, from the detector response, computed with the pulse-height (F8) tally. Although experimental validation on Orano CIME calibration blocks demonstrates good agreement, this approach introduces a significant spectral distortion when simulating realistic water-filled boreholes. To overcome this limitation, we developed a Gaussian process (GP) surrogate model trained to emulate analog F8 simulations, which outperforms the two-step approach in both computing time and accuracy when reproducing gamma-ray spectra from an experimental log.

Förderung

  • Research and Development Collaboration between Commissariat à l'Energie Atomique et aux énergies alternatives (CEA) and Orano Mining

Zitieren

APA

Pellet-Rostaing, A., Pérot, B., Marchais, T., Pérot, N., Toubon, H., Bensedik, Y., Hocquet, S., Pouet, C., & Fouladirad, M. (2026). Fast Generation of Gamma-Ray Logs Based on MCNP Simulations for Uranium Mining. IEEE Transactions on Nuclear Science, 73(6), 2224-2231. https://doi.org/10.1109/tns.2026.3672048

Vancouver

Pellet-Rostaing A, Pérot B, Marchais T, Pérot N, Toubon H, Bensedik Y, et al. Fast Generation of Gamma-Ray Logs Based on MCNP Simulations for Uranium Mining. IEEE Transactions on Nuclear Science. 2026;73(6):2224-2231. doi:10.1109/tns.2026.3672048

BibTeX
@article{PelletRostaing2026,
  author = {Pellet-Rostaing, Arthur and Pérot, Bertrand and Marchais, Thomas and Pérot, Nadia and Toubon, Hervé and Bensedik, Youcef and Hocquet, Sebastien and Pouet, Christophe and Fouladirad, Mitra},
  title = {Fast Generation of Gamma-Ray Logs Based on MCNP Simulations for Uranium Mining},
  journal = {IEEE Transactions on Nuclear Science},
  year = {2026},
  volume = {73},
  number = {6},
  pages = {2224--2231},
  doi = {10.1109/tns.2026.3672048}
}
Kostenlos starten →