American Nuclear Society
Home

Home / Publications / Journals / Nuclear Technology / Volume 193 / Number 1

RelapManager: A Safety Analysis Automation Framework and Software for RELAP5-3D

Robert P. Martin, Simone H. Morgan, Christopher Cragg

Nuclear Technology / Volume 193 / Number 1 / January 2016 / Pages 200-212

Technical Paper / Special Issue on the RELAP5-3D Computer Code / dx.doi.org/10.13182/NT14-140

First Online Publication:November 30, 2015
Updated:January 4, 2016

Evaluation methodologies are under development at BWX Technologies (BWXT) to support licensure of the BWXT small modular reactor. In these evaluation methodologies, RELAP5-3D is being used as the computation engine for reactor coolant system performance. To facilitate analyses, the RelapManager code was developed as a Windows-based graphical user interface for RELAP5-3D. It is used to edit RELAP5-3D input models and develop simulation suites with unique attention to the preparation of Quality Program records required per 10 CFR 50 Appendix B. The software has been created with Visual Studio employing the .NET integrated development environment. The programming focal point for RelapManager is an XML schema that captures the modeling rules inherent in RELAP5-3D. It transforms an existing input model into XML, which, through a large library of functions created to support XML file types, provides a rich user experience capable of improving productivity and quality.

RelapManager includes specialized features for the preparation of uncertainty and sensitivity/importance analyses, execution on a remote server, plotting, and data checking against the modeling rules. As a demonstration of these features, an uncertainty and importance analysis has been prepared using the “Typical PWR” test problem provided with the nominal RELAP5-3D release package. RelapManager takes as input uncertainty models characterizing the expected domain of model parameter variability. Subsequently, a calculation suite derived from sampling the uncertainty domain is described, and a run script is prepared by RelapManager. Figure of merit results are gathered by RelapManager and analyzed using a variance-based sensitivity analysis method.