BEPU Analysis and Benchmark with IIST 2% SBLOCA Experiment Using TRACE/DAKOTA (NUREG/IA-0456)

On this page:

Download complete document

Publication Information

Manuscript Completed: April 2015
Date Published: September 2015

Prepared by:
Chunkuan Shih*, Jung-Hua Yang*, Jong-Rong Wang*, Hao-Tzu Lin, Show-Chyuan Chiang**,
Chia-Chuan Liu**

Institute of Nuclear Energy Research, Atomic Energy Council, R.O.C.
1000, Wenhua Rd., Chiaan Village, Lungtan, Taoyuan, 325, Taiwan

*Institute of Nuclear Engineering and Science, National Tsing Hua University
101 Section 2, Kuang Fu Rd., HsinChu, Taiwan

**Department of Nuclear Safety, Taiwan Power Company
242, Section 3, Roosevelt Rd., Zhongzheng District, Taipei, Taiwan

K. Tien, NRC Project Manager

Division of Systems Analysis
Office of Nuclear Regulatory Research
U.S. Nuclear Regulatory Commission
Washington, DC 20555-0001

Prepared as part of:
The Agreement on Research Participation and Technical Exchange
Under the Thermal-Hydraulic Code Applications and Maintenance Program (CAMP)

Published by:
U.S. Nuclear Regulatory Commission
Washington, DC 20555-0001

Availability Notice


There are two licensing approaches for evaluation of Loss of Coolant Accidents consequences: conservative methodology and best estimate methodology. Recently, the trend of nuclear reactor safety analysis reveals an increasing interest to substitute best estimate for conservative methodologies to achieve the safety margins and regulate the licensing and operations of nuclear reactors.

The Institute of Nuclear Energy Research Integral System Test (IIST) facility is a test facility to simulate the thermal hydraulics of a Westinghouse 3-loop Pressurized Water Reactor at Maanshan Nuclear Power Plant. The research purposes of the IIST facility are: (a) to enhance the understanding of thermal hydraulics during transients as well as Small Break Loss of Coolant Accidents, (b) to contribute to the evaluations and developments of safety computer codes, (c) to validate the Emergency Operation Procedures during the transients. The scaling factors of the IIST facility for height and volume in the Reactor Coolant System are approximately 1/4 and 1/400, respectively, and the maximum operating pressure is 2.1 MPa. The scaling of hot leg is based on the Froude number criterion to simulate the transition of flow regimes in the horizontal pipes during transients and accidents.

This study is developed the BEPU methodology and the uncertainty results were compared with the IIST experiment data. The IIST TRACE model consists of 89 hydraulic components, 243 control blocks, 39 heat structures and a power component. The data interactions and communications between TRACE and DAKOTA were controlled by SNAP. Finally, correlations between input parameters and output data are calculated for sensitivity study and ranking to investigate what input parameters dominate the contribution of uncertain distribution of PCT.

There are two tasks in this study. One is benchmark of the simulation capability of IIST TRACE model by comparing with the IIST SBLOCA experiment results. The other is BEPU in SBLOCA analysis, several important parameters were considered in the uncertainty quantified. The BEPU analysis is focused on the discussion of measurement uncertainty and model uncertainty. Furthermore, the IIST experimental data were used to benchmark the results of uncertainty analysis. There are 5 parameters taken into account in this uncertainty analysis. An uncertainty band is formed by the 59 calculations, and the benchmark results show the IIST experimental data were located in the uncertainty band. In addition, this study used two methods to calculate correlation coefficients between an input and an output variable: Pearson’s correlation coefficient and Spearman’s rank correlation. The formula quantifies the correlations between the input and output parameter. Even though the Spearman’s rank correlation employs the rank data is difference than the Pearson’s correlation, their results are similar trend. The results indicate that choked flow coefficient is the most sensitive parameter. The correlation coefficients of Pearson’s and Spearman’s are -0.72405 and -0.70199 respectively. Although the measurement error may influence the PCT calculated, its effect is smaller than the model uncertainty.

Page Last Reviewed/Updated Tuesday, March 09, 2021