Theoretical analysis of neutron spectra measurement with SiC detectors using a machine learning technique

Abstract Detecting neutrons is crucial for various nuclear applications. Currently, there exists a technology gap in detecting fast neutrons under high-flux conditions, primarily due to some limitations solid detectors face in supporting high flux. Addressing these challenges requires a balanced approach that combines a detector capable of providing exploitable signals with an effective method for reconstructing the neutron spectrum. This work explores the potential of utilizing machine learning techniques to address this gap by determining neutron spectra using SiC detectors. The machine learning algorithm described in this paper has been tested on measured spectra stored in the IAEA Compendium database. The outcomes of this theoretical study demonstrate the feasibility of the proposed approach in determining neutron spectra from various neutron sources. However, experimental validation is necessary to fully validate the methodology and ensure its applicability in practical settings.