STOCHASTIC AND DATA UNCERTAINTY QUANTIFICATION OF MCNP
PREDICTED NUCLIDE CONCENTRATIONS IN FUEL BURNUP SIMULATIONS

Year
2023
Author(s)
Seray Cerezo - Department of Nuclear Engineering, Texas A&M University, College Station
Grace Long - Texas A&M University
Sunil S. Chirayath - Dept of Nuclear Engineering, Texas A&M University
Corey Keith - Los Alamos National Laboratory
File Attachment
Abstract
Monte Carlo N-Particle transport code (MCNP) is used widely to simulate nuclear fuel burnup and depletion because it is efficient in solving the radiation transport equation for complex geometries. MCNP simulates fuel burnup and estimates the concentrations of actinides and fission products, which are useful in nuclear safeguards monitoring. The MCNP fuel burnup simulation does not estimate the uncertainties in the predicted nuclide concentration caused due to the uncertainties in nuclear data and the stochastic radiation transport methodology used. The nuclide concentration is calculated through CINDER 90 nuclide generation and depletion module, which uses the neutron reaction rates and flux values calculated by MCNP. Stochastic uncertainties in the neutron reaction rates and flux values are calculated by MCNP, which introduces stochastic uncertainty in the nuclide concentration, but this uncertainty is not propagated through each burnup time step to estimate the uncertainty in the nuclide concentrations predicted. This uncertainty is in addition to the systematic uncertainty caused due to the nuclear data. The neutron reaction rates can be broken down into neutron flux, number density, and microscopic neutron interaction cross section terms. The number density and neutron flux are provided by MCNP, and the neutron flux term calculated will contain stochastic uncertainty; however, the microscopic cross sections used in MCNP will contain systematic uncertainties. Propagating the effects of both sources of uncertainties using a Backward Euler numerical scheme allows for the reporting of the total relative error in the predicted nuclide concentrations. The MCNP depletion calculation uses a one group neutron flux and therefore a one group microscopic neutron cross section is necessary to find the neutron reaction rates. The microscopic neutron cross sections are dependent on the energy spectrum of the flux in the fuel burnup simulation. The process of acquiring these microscopic cross sections and weighting them by the flux is automated in our study for estimating both stochastic and systematic uncertainties. The final product of the study will be a software that calculates and reports stochastic, systematic, and total nuclide concentration uncertainties by utilizing the MCNP fuel burnup simulation output file.