Nuclear Engineering and Technology

Korean Nuclear Society (한국원자력학회)
권호 표지
DOI QR코드

DOI QR Code

An adaptive deviation-resistant neutron spectrum unfolding method based on transfer learning

  • Cao, Chenglong (Key Laboratory of Neutronics and Radiation Safety, Anhui Province Key Laboratory of Neutron Physics and Control Technology, Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences) ;
  • Gan, Quan (Key Laboratory of Neutronics and Radiation Safety, Anhui Province Key Laboratory of Neutron Physics and Control Technology, Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences) ;
  • Song, Jing (Key Laboratory of Neutronics and Radiation Safety, Anhui Province Key Laboratory of Neutron Physics and Control Technology, Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences) ;
  • Yang, Qi (Key Laboratory of Neutronics and Radiation Safety, Anhui Province Key Laboratory of Neutron Physics and Control Technology, Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences) ;
  • Hu, Liqin (Key Laboratory of Neutronics and Radiation Safety, Anhui Province Key Laboratory of Neutron Physics and Control Technology, Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences) ;
  • Wang, Fang (Key Laboratory of Neutronics and Radiation Safety, Anhui Province Key Laboratory of Neutron Physics and Control Technology, Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences) ;
  • Zhou, Tao (Key Laboratory of Neutronics and Radiation Safety, Anhui Province Key Laboratory of Neutron Physics and Control Technology, Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences)
  • Received : 2019.12.10
  • Accepted : 2020.04.27
  • Published : 2020.11.25
Download PDF
⟨ Previous Next ⟩

Abstract

Neutron spectrum is essential to the safe operation of reactors. Traditional online neutron spectrum measurement methods still have room to improve accuracy for the application cases of wide energy range. From the application of artificial neural network (ANN) algorithm in spectrum unfolding, its accuracy is difficult to be improved for lacking of enough effective training data. In this paper, an adaptive deviation-resistant neutron spectrum unfolding method based on transfer learning was developed. The model of ANN was trained with thousands of neutron spectra generated with Monte Carlo transport calculation to construct a coarse-grained unfolded spectrum. In order to improve the accuracy of the unfolded spectrum, results of the previous ANN model combined with some specific eigenvalues of the current system were put into the dataset for training the deeper ANN model, and fine-grained unfolded spectrum could be achieved through the deeper ANN model. The method could realize accurate spectrum unfolding while maintaining universality, combined with detectors covering wide energy range, it could improve the accuracy of spectrum measurement methods for wide energy range. This method was verified with a fast neutron reactor BN-600. The mean square error (MSE), average relative deviation (ARD) and spectrum quality (Qs) were selected to evaluate the final results and they all demonstrated that the developed method was much more precise than traditional spectrum unfolding methods.

Keywords

References

  1. J.M. Ortiz-Rodriguez, A.R. Alfaro, A.R. Haro, et al., A neutron spectrum unfolding computer code based on artificial neural networks, Radiat. Phys. Chem. 95 (4) (2014) 428-431. https://doi.org/10.1016/j.radphyschem.2013.05.007
  2. J. Wang, Y. Zhou, et al., Neutron spectrum unfolding using three artificial intelligence optimization methods, Appl. Radiat. Isot. 147 (2019) 136-143, 2019. https://doi.org/10.1016/j.apradiso.2019.03.009
  3. H. Xiong, Research of On-Line Neutron Spectrum Measurement Method for Nuclear Reactor, University of Science and Technology of China, 2018.
  4. R. Bedogni, C. Domingo, A. Esposito, et al., FRUIT: an operational tool for multisphere neutron spectrometry in workplaces, Nucl. Instrum. Methods Phys. Res. A 580 (3) (2007) 1301-1309. https://doi.org/10.1016/j.nima.2007.07.033
  5. J. Shore, R. Johnson, Axiomatic derivation of the principle of maximum entropy and the principle of minimum cross-entropy, IEEE Trans. Inf. Theor. 26 (1) (1980) 26-37. https://doi.org/10.1109/TIT.1980.1056144
  6. M. Matzke, Unfolding procedures, Radiat. Protect. Dosim. 107 (1-3) (2003) 155-174. https://doi.org/10.1093/oxfordjournals.rpd.a006384
  7. E.T. Jaynes, Prior information and ambiguity in inverse problems, Inverse Problems, 1984, p. 14.
  8. R. Sanna, K. O"Brien, Monte-Carlo unfolding of neutron spectra, Nucl. Instrum. Methods 91 (4) (1971) 573-576. https://doi.org/10.1016/0029-554X(71)90680-X
  9. Vega-Carrillo Hector Rene, Hernandez-Davila Victor Martin, et al., Neutron spectrometry using artificial neural networks, Radiat. Meas. 41 (4) (2006) 425-431. https://doi.org/10.1016/j.radmeas.2005.10.003
  10. N. Mohammadi, H.M. Hakimabad, L.R. Motavalli, Neural network unfolding of neutron spectrum measured by gold foil-based Bonner sphere, J. Radioanal. Nucl. Chem. 303 (3) (2015) 1687-1693.
  11. H. Shahabinejad, S.A. Hosseini, M. Sohrabpour, A new neutron energy spectrum unfolding code using a two steps genetic algorithm, Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 811 (2016) 82-93. https://doi.org/10.1016/j.nima.2015.12.028
  12. Abolfazl S. Hosseini, Neutron spectrum unfolding using artificial neural network and modified least square method, Radiat. Phys. Chem. 126 (2016) 75-84. https://doi.org/10.1016/j.radphyschem.2016.05.010
  13. A.A. Alvar, M.R. Deevband, M. Ashtiyani, Neutron spectrum unfolding using radial basis function neural networks, Appl. Radiat. Isot. (2017), S0969804316308259.
  14. C.C. Braga, M.S. Dias, Application of Neural Networks for unfolding neutron spectra measured by means of Bonner Spheres, Nucl. Instrum. Methods Phys. Res. 476 (1-2) (2002) 252-255. https://doi.org/10.1016/S0168-9002(01)01464-4
  15. D.W. Freeman, D.R. Edwards, A.E. Bolon, Genetic algorithms-A new technique for solving the neutron spectrum unfolding problem, Nucl. Instrum. Methods Phys. Res., Sect. A 425 (3) (1999) 549-576. https://doi.org/10.1016/S0168-9002(98)01427-2
  16. K.P. Sudheer, A.K. Gosain, D.M. Rangan, et al., Modeling evaporation using an artificial neural network algorithm, Hydrol.Processes 16 (2002) 3189-3202. https://doi.org/10.1002/hyp.1096
  17. R. Kardan M, S. Setayeshi, R. Koohi-Fayegh, et al., Neutron spectra unfolding in Bonner spheres spectrometry using neural networks, Radiat. Protect. Dosim. 104 (1) (2003) 27-30. https://doi.org/10.1093/oxfordjournals.rpd.a006158
  18. X. Yin, J. Han, J. Yang, et al., Efficient classification across multiple database relations: a CrossMine approach, IEEE Trans. Knowl. Data Eng. 18 (6) (2006) 770-783783. https://doi.org/10.1109/TKDE.2006.94
  19. L.I. Kuncheva, J.J. Rodriguez, Classifier ensembles with a random linear oracle, IEEE Trans. Knowl. Data Eng. 19 (4) (2007) 500-508. https://doi.org/10.1109/TKDE.2007.1016
  20. S.J. Pan, Q. Yang, A survey on transfer learning, IEEE Trans. Knowl. Data Eng. 22 (10) (2010) 1345-1359. https://doi.org/10.1109/TKDE.2009.191
  21. C. Cao, P. Long, Q. Gan, et al., Research on online measurement method of wide range for reactor based on spectral features recognition, Nucl. Sci. Technol. (4) (2019).
  22. Y. Wu, J. Song, H. Zheng, et al., CAD-based Monte Carlo program for integrated simulation of nuclear system SuperMC, Ann. Nucl. Energy 82 (2015) 161-168. https://doi.org/10.1016/j.anucene.2014.08.058
  23. Y. Wu, Multi-functional neutronics calculation methodology and program for nuclear design and radiation safety evaluation, Fusion Sci. Technol. 74 (4) (2018) 321-329. https://doi.org/10.1080/15361055.2018.1475162
  24. Y. Wu, Fds Team, CAD-based interface programs for fusion neutron transport simulation, Fusion Eng. Des. 84 (7-11) (2009) 1987-1992. https://doi.org/10.1016/j.fusengdes.2008.12.041
  25. R.V. Griffith, J. Palfalvi, U. Madh-Vanath, Compendium of Neutron Spectra and Detector Responses for Radiation Protection purpose[R], International-al Atomic Energy Agency, Vienna, 1990.
  26. Q. Zhu, L. Tian, X. Yang, et al., Advantages of artificial neural network in neutron spectra unfolding, Chin. Phys. Lett. : English version 31 (7) (2014) 69-72.
  27. D. Stuenkel, J.P. Holloway, G.F. Knoll, Neutron spectrum unfolding using a modified truncated singular value decomposition method, Nucl. Sci. Eng. 132 (3) (1999) 261-272. https://doi.org/10.13182/NSE99-A2062
  28. Iaea, BN-600 Hybrid Core Benchmark analyses[R], IAEA-TECDOC-1632, Vienna, 2010.
  29. C. Cao, Q. Gan, J. Song, et al., A two-step neutron spectrum unfolding method for fission reactors based on artificial neural network, Ann. Nucl. Energy 139 (2020).

Cited by

  1. A novel method for rapid calculation of moderated neutron spectrum and its application in deep penetration vol.168, 2020, https://doi.org/10.1016/j.anucene.2021.108895