Review of Spherical Neutron Spectrometers and AI Innovations in Neutron Spectrum Unfolding
ID:59 View Protection:ATTENDEE Updated Time:2024-09-23 21:56:13 Hits:317 Oral Presentation

Start Time:Pending(Asia/Shanghai)

Duration:Pending

Session:No Session »

Presentation File

Tips: Only the registered participant can access the file. Please sign in first.

Abstract
Neutron spectrometry is a technique used to measure the neutron energy distribution covering a wide range of energies up to several MeV. In terms of radiation protection, neutron survey meters are usually deployed for radiation monitoring. However, neutron survey meters have limitations when used for dose measurements, so they are typically supplemented with neutron spectrometry. Among the various methods used for neutron spectrometry is the Bonner sphere spectrometer. It consists of a thermal neutron detector located in the center of polyethylene spheres with different diameters. This design tends to be bulky, heavy and difficult to use in field measurements, therefore several studies have been made to design a spectrometer that is lighter and more compact. This review details the alternative designs of neutron sphere spectrometers, such as single sphere spectrometer with water as moderator, portable single sphere neutron spectrometer based on glass scintillators that is tested in neutron field, and water moderator nested spheres neutron spectrometer. In addition, this review also focuses on multi shell single sphere moderating assembly with OSL detector pairs and multi shell neutron spectrometer with passive and active neutron detectors. One of the recent developments for neutron spectrometry is the use of computer unfolding methods for determining neutron spectra. Advancement in neutron unfolding techniques, particularly through the development of algorithms utilizing artificial intelligence (AI), has significantly affected the field of neutron spectroscopy. This review further discusses the integration of AI optimization methods, such as neural networks, regression methods and genetic algorithms. AI algorithms can be constructed from huge datasets of simulated and experimental neutron spectra, not only optimizing the unfolding process to eliminate errors from sphere readings but also accelerates data analysis.
Keywords
neutron spectrometry, neutron detector, scintillators, spectrometer, unfolding
Speaker
Marinell Palangao
PhD student Harbin Engineering University

Submission Author
Marinell Palangao Harbin Engineering University
Yushou Song Harbin Engineering University
Shengqiang Chen Harbin Engineering University
Submit Comment
Verify Code Change Another
All Comments
Important Date
  • Conference Date

    Sep 23

    2024

    to

    Sep 25

    2024

  • Sep 24 2024

    Contribution Submission Deadline

  • Sep 25 2024

    Registration deadline

Sponsored By
Harbin Engineering University (HEU)
Contact Information