Search NASA⌕ Search

Engineering topics

Padovani, Enrico

Publications and source records attributed to Padovani, Enrico.

3 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

Validation of MCNPX-PoliMi code for simulations of radioxenon beta-gamma coincidence detection

Radioxenon detection is an important component of the verification regime for the Comprehensive Nuclear-Test-Ban Treaty. We developed and validated a new model in MCNPX-PoliMi to simulate the decay of various radioxenon isotopes and the detector response of a variety of detector types. The model was validated using calibration data from a plastic and NaI(Tl) beta-gamma coincidence detector and the results are presented. The results of this validation show that this model can also be used as a tool to produce training spectra and as a method to calibrate radioxenon detection systems.

Radioxenon analysis, MCNP, beta-gamma coincidence↗
DOE OSTI2015

An Implicit Correlation Method for Cross-Correlation Sampling, with MCNPX-PoliMi Validation

Monte Carlo particle transport codes used to model detector response are traditionally run in analog mode. However, analog simulations of cross-correlation measurements are extremely time-consuming because the probability of coincident detection is small, approximately equal to the product of the probabilities of a single detection in each detector. The new “implicit correlation” method described here increases the number of correlated event scores, thereby reducing variance and required computation times. The cost of the implicit correlation method is comparable to the cost of simulating single-event detection for the lowest absolute detector efficiency in the problem. The new method is especially useful in the nuclear nonproliferation and safeguards fields for simulating correlation measurements of shielded special nuclear material. The new method was implemented in MCNPX-PoliMi for neutron-neutron cross-correlations with a Cf-252 spontaneous fission source measured by 14 detectors at various angles. The method demonstrated good agreement with analog simulation and reference measurement results. Small differences between nonanalog and analog cross-correlation distributions are attributed to discretization errors that are often not present in practical applications. Improvement in the figure of merit was greater than a factor of 100 in all tested cases.

Marcath, Matthew J.↗