skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Statistics and integral experiments in the verification of LOCA calculations models. [BWR; PWR]

Conference ·
OSTI ID:6555786

The LOCA (loss of coolant accident) is a hypothesized, low-probability accident used as a licensing basis for nuclear power plants. Computer codes which have been under development for at least a decade have been the principal tools used to assess the consequences of the hypothesized LOCA. Models exist in two versions. In EM's (Evaluation Models) the basic engineering calculations are constrained by a detailed set of assumptions spelled out in the Code of Federal Regulations (10 CFR 50, Appendix K). In BE Models (Best Estimate Models) the calculations are based on fundamental physical laws and available empirical correlations. Evaluation models are intended to have a pessimistic bias; Best Estimate Models are intended to be unbiased. Because evaluation models play a key role in reactor licensing, they must be conservative. A long-sought objective has been to assess this conservatism by combining Best Estimate Models with statisticallly established error bounds, based on experiment. Within the last few years, an extensive international program of LOCA experiments has been established to provide the needed data. This program has already produced millions of measurements of temperature, density, and flow and millions of more measurements are yet to come.

Research Organization:
EG and G Idaho, Inc., Idaho Falls (USA)
DOE Contract Number:
EY-76-C-07-1570
OSTI ID:
6555786
Report Number(s):
CONF-781108-1; TRN: 79-002019
Resource Relation:
Conference: 4. DOE statistical symposium, Albuquerque, NM, USA, 1 Nov 1978
Country of Publication:
United States
Language:
English