American Nuclear Society
Home

Home / Store / Journals / Electronic Articles / Nuclear Science and Engineering / Volume 175 / Number 3 / Pages 292-307

Experimental Validation of 3-D Reconstructed Pin Power Distributions in Full-Scale BWR Fuel Assemblies with Partial-Length Rods

Flavio Dante Giust, Peter Grimm, Rakesh Chawla

Nuclear Science and Engineering / Volume 175 / Number 3 / Pages 292-307

November 2013

Format:

Price:$30.00
Member Price:$27.00
Member Savings:$3.00

Total fission rate measurements have been performed on full-size boiling water reactor fuel assemblies of type SVEA-96 Optima2 in the framework of phase III of the light water reactor (LWR)-PROTEUS experimental program at Paul Scherrer Institute. This paper presents comparisons of calculated, nodal reconstructed, pinwise total fission rate distributions with experimental results. Radial comparisons have been performed for the three axial sections of the assembly (96, 92, and 84 fuel pins), while three-dimensional (3-D) effects have been investigated at pellet level for the two transition regions, i.e., the tips of the short (one-third) and long (two-thirds) partial-length rods. The test zone has been modeled using two different code systems: HELIOS/PRESTO-2 and CASMO-5/SIMULATE-5. The former is presently used for core monitoring and design at the Leibstadt Nuclear Power Plant (KKL). The latter represents the most recent generation of codes constituting the widely applied CASMO/SIMULATE system. For representing the PROTEUS test zone boundaries, partial current ratios - derived from a 3-D Monte Carlo (MCNPX) model of the entire reactor - have been applied to the PRESTO-2 and SIMULATE-5 models in the form of two-group and five-group diagonal albedo matrices, respectively. The MCNPX results have also served as a reference high-order transport solution in the calculation-to-experiment (C/E) comparisons.

It is shown that the performance of the nodal methodologies in predicting the global distribution of the total fission rate is very satisfactory. Considering the various radial comparisons, the standard deviations of the C/E distributions do not exceed 1.9% for any of the three methodologies - PRESTO-2, SIMULATE-5, and MCNPX. For the 3-D comparisons at pellet level, the corresponding standard deviations are 2.7%, 2.0%, and 2.1%, respectively.

 
 
 
Questions or comments about the site? Contact the ANS Webmaster.
advertisement