Validation of Monte-Carlo methods for generation time and delayed neutron fraction predictions

Jan Wagemans, Peter Baeten, Anatoly Kochetkov, Grégory Perret, Patrick D. Blaise, Nadia Messaoudi, Benoit Geslot, Andreas Pautz

Research outputpeer-review


The capability of Monte-Carlo codes to predict kinetic parameters of nuclear systems is validated against a series of experiments in zero-power reactors. Experimental data are issued from facilities operated by the CEA, SCK.CEN and PSI research institutes and analyzed in the framework of the Venus–Eole–Proteus international collaboration. Facilities were configured to study several reactor types (high temperature reactors, light water reactors, material testing reactors, accelerator driven system demonstrators) and type of spectra (thermal, epithermal and fast). Monte-Carlo codes are used to predict the effective generation time and in some cases the effective delayed neutron fraction. The benchmarked codes are MCNP5 and MCNPX coupled to the LAMBDA scripts developed at SCK.CEN. Generation time predictions from the two codes agree within 2.5% for values larger than 1 s but have larger discrepancies (up to 7%) for faster systems. Discrepancies with the measured values depend largely on the selected experiment and can reach up to 9%. Delayed neutron predictions with MCNP5 compare well (3–4%) with all measurements.
Original languageEnglish
Pages (from-to)165-170
Number of pages6
JournalAnnals of nuclear energy
StatePublished - 1 Nov 2016

Cite this