Neutron spectrum unfolding using two architectures of convolutional neural networks
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
We deploy artificial neural networks to unfold neutron spectra from measured energy-integrated quantities. These neutron spectra represent an important parameter allowing to compute the absorbed dose and the kerma to serve radiation protection in addition to nuclear safety. The built architectures are inspired from convolutional neural networks. The first architecture is made up of residual transposed convolution's blocks while the second is a modified version of the U-net architecture. A large and balanced dataset is simulated following “realistic” physical constraints to train the architectures in an efficient way. Results show a high accuracy prediction of neutron spectra ranging from thermal up to fast spectrum. The dataset processing, the attention paid to performances' metrics and the hyperoptimization are behind the architectures' robustness.
Abstract
We deploy artificial neural networks to unfold neutron spectra from measured energy-integrated quantities. These neutron spectra represent an important parameter allowing to compute the absorbed dose and the kerma to serve radiation protection in addition to nuclear safety. The built architectures are inspired from convolutional neural networks. The first architecture is made up of residual transposed convolution's blocks while the second is a modified version of the U-net architecture. A large and balanced dataset is simulated following “realistic” physical constraints to train the architectures in an efficient way. Results show a high accuracy prediction of neutron spectra ranging from thermal up to fast spectrum. The dataset processing, the attention paid to performances' metrics and the hyperoptimization are behind the architectures' robustness.
참고문헌 (15)
- M. Reginatto, "Bayesian approach for quantifying the uncertainty of neutron doses derived from spectrometric measurements", Radiat. Protect. Dosim, 121(1), pp.64~69 (2006)
- S. Itoh, "A fundamental study of neutron spectra unfolding based on the maximum likelihood method", Nucl. Instrum. Methods A, 251(1), pp.144~155 (1986)
- M. Reginatto, "Overview of spectral unfolding techniques and uncertainty estimation", Radiat. Meas, 45(10), pp.1323~1329 (2010)
- H. N. Mülthei, "On an iterative method for the unfolding of spectra", Nucl. Instrum. Methods A, 257(2), pp.371~377 (1987)
- Chenglong Cao, "An adaptive deviation-resistant neutron spectrum unfolding method based on transfer learning", Nuclear Engineering and Technology, 52(11), pp.2452~2459 (2020), 한국원자력학회[DOI]
- S. A. Hosseini, "Neutron spectrum unfolding using artificial neural network and modified least square method", Radiat. Phys. Chem, 126, pp.75~84 (2016)
- C. Cao, "A two-step neutron spectrum unfolding method for fission reactors based on artificial neural network", Ann. Nucl. Energy, 139, pp.107219~ (2020)
- J. Wang, "Neutron spectrum unfolding based on generalized regression neural networks for neutron fluence and neutron ambient dose equivalent estimations", Appl. Radiat. Isot, 154, pp.108856~ (2019)
- A. A. Alvar, "Neutron spectrum unfolding using radial basis function neural networks", Appl. Radiat. Isot, 129, pp.35~41 (2017)
- R. Yamashita, "Convolutional neural networks : an overview and application in radiology", Insights Imaging, 9, pp.611~629 (2018)
- D. Im, "D.T.- CNN, Dilated and transposed convolution neural network accelerator for real-time image segmentation on mobile devices", Proc. Of the IEEE International Symposium of Circuits and Systems (ISCAS), pp.1~5 (2019)
- O. Ronneberger, "Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015" (2015), Springer
- Maha Bouhadida, "Neutron spectra reconstruction based on an artificial neural network trained with a large built dataset", ICRS 14, 2022, ~[DOI]
- F. Trompier, "Dosimetry of the mixed field irradiation facility CALIBAN", Radiat. Meas, 43, pp.1077~1080 (2008)
- T. Akiba, "Optuna: a next-generation hyperparameter optimization framework", Proc. Of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp.2623~2631 (2019)