The New Nuclear Data Sensitivity Analysis and Uncertainty Propagation Tool in NESTLE

Show full item record

Title: The New Nuclear Data Sensitivity Analysis and Uncertainty Propagation Tool in NESTLE
Author: Wieselquist, William Adam
Advisors: D.Y. Anistratov, Committee Chair
Paul J. Turinsky, Committee Member
Mansoor Haider, Committee Member
Abstract: In support of the need for better design and evaluation tools for reactor-based transmutation systems we have upgraded NESTLE, the 2/4 energy group thermal reactor physics code of the Nuclear Engineering Department at North Carolina State University with: i) the ability to perform nuclide transmutation calculations for a general, user-defined field of nuclei and transmutation paths and ii) the ability to analyze sensitivities and propagate uncertainties in the end-of-cycle (EOC) nuclide inventory with respect to nuclear data and beginning-of-cycle (BOC) nuclide inventory. We present two methods of sensitivity analysis: i) direct perturbation and recalculation (DPAR) and ii) sensitivity analysis utilizing an adjoint system (AS). With DPAR, we simply perturb data and recalculate solutions of our system and thus may analyze sensitivity of all responses to perturbations in one data parameter per solution of the perturbed forward problem. With the AS, we form a system of equations, the solution of which may be used to estimate the first variation of a response with respect to any data parameters. For the AS, we have developed the equations for both the predictor and predictor-corrector neutron/nuclide field coupling methods in NESTLE. To our knowledge, the AS for the predictor-corrector coupling has never been presented. Then we used the tools we have developed to evaluate the sensitivity of EOC nuclide concentrations and SNF hazard measures with respect to nuclear data for a cycle 1 pressurized water reactor (PWR) core. In our study, we found that the nuclear data crucial to modeling US reactors' once-through cycle (fission cross sections of 235U and 239Pu, the main fuel nuclei, and capture cross sections for 238U) also has the highest impact on EOC nuclide inventory of so-called "problem nuclei" (e.g. Am, Cm, etc.) Note that these results only apply to cycle 1, in which fresh fuel is irradiated for the first time. Because most fuel assemblies are present in the core for three cycles, two more cycles should be considered to analyze sensitivity of spent nuclear fuel (SNF)---the fuel for transmutation systems---with respect to nuclear data. Although, NESTLE has the capability to perform sensitivity analysis over multiple cycles, NESTLE does not have the capability to determine loading patterns and the benchmark case we used did not include cycle 2 or cycle 3 data.
Date: 2005-05-02
Degree: MS
Discipline: Nuclear Engineering

Files in this item

Files Size Format View
etd.pdf 1.377Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record