Published on the occasion of its 70th General Conference (Vienna, September 14–18, 2026), the International Atomic Energy Agency's (IAEA) Nuclear Technology Review 2026 provides an overview of the most significant advancements in nuclear technologies. This year, two topics are dedicated to instrumentation, highlighting research on artificial intelligence-assisted neutron spectrometry conducted by the Laboratoire national Henri Becquerel (LNE-LNHB).
An approach tailored to complex radiation fields
The strength of this approach? Its ability to characterize neutron spectra with high precision in complex radiation fields, where fast neutrons, thermal neutrons, and intense photon emissions coexist.
To disentangle these signals and analyze complex neutron spectra, the laboratory's researchers combine several technological and methodological building blocks:
Organic scintillators capable of discriminating neutrons from photons;
Artificial intelligence trained on both experimental and simulated data;
Monte Carlo simulations to quantify uncertainties.
The objective: to identify specific signatures and, more broadly, to reconstruct plausible, robust, and interpretable neutron spectra.
Serving the metrology of ionizing radiation
Image
Already documented in several scientific publications, this research directly contributes to the mission entrusted by the LNE to the Laboratoire national Henri Becquerel: the development of precisely characterized new reference radiation fields.
