Neural ray tracing for a neutron triple axis spectrometer
No Thumbnail Available
Date
2023-03
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
American Physical Society
Abstract
The behaviour of a neutron scattering instrument may be approached both analytically as well as numerically. Traditionally numerical studies entail some form of ray-tracing approach that allows for a realistic and accurate model of the behaviour of an instrument. In-particular resolution and flux are two important parameters that require accurate determination when building an instrument like a triple-axis. The numerical approach produces realistic results albeit often using a skeleton type model of the instrument. Despite the success of a triple-axis configuration where radiation is conditioned in reciprocal and energy space, background signal observed in all spectra measured, is a very important parameter that isn’t easily quantifiable without including the structural and shielding materials that also go into a full working spectrometer. © 2025 American Physical Society
The advent of deep learning via simulated neural networks is a significant development that we wish to explore to describe the behaviour of a triple axis spectrometer when thermal neutron radiation passes through it. Neural representations of the solid model and neutron transport are rendered to produce radiation maps S(e,r) at any point along the instrument and in the area surrounding it
Description
Keywords
Spectrometers, Scattering, Measuring instruments, Spectra, Shielding materials, Neutron radiography
Citation
Stampfl, A. P. (2023). Neural ray tracing for a neutron triple axis spectrometer. Poster presented to the 2023 APS March Meeting, Las Vegas, Nevada (March 5-10), Virtual (March 20-22). In Bulletin of the American Physical Society, 68(3), VV01.00054. Retrieved from: https://meetings.aps.org/Meeting/MAR23/Session/VV01.54.