Advances in fracture mechanics analyses of primary system performance under operating and accident conditions

Abstract

Listen

Translate article

Safety research sponsored by the Nuclear Regulatory Commission, Division of Reactor Safety Research, has resulted in notable advances in several areas of importance in the safety evaluation of reactor primary systems under normal operations and accident situations. First, the methods of linear elastic fracture mechanics and of elastic plastic fracture mechanics have been validated for prediction of pressure vessel performance by the Intermediate Vessel Test program results at the Oak Ridge National Laboratory. The test results from both hydraulically and pneumatically loaded pressure vessels are in good agreement with the pretest predictions of vessel failure for the given conditions of pressure, temperature, material properties, and type and size of flaw. This has been verified in base material, both plate and forging, in submerged arc, stress relieved welds, and in weld repair regions. This ability confidently to predict vessel performance under realistic service conditions has permitted development of the computer program OCTAVIA which computes failure curves for a range of flaw sizes in terms of pressure and temperature for specified pressure vessel material at specific neutron fluence levels. It then considers the probability of occurrence of flaw sizes and magnitude of pressure during an operational, overpressurization transient and determines the probability of failure, for both individual flaw sizes and for the full spectrum. A second advance in fracture mechanics technology is related to thermal shock analysis. This advance has been verified by the confirmatory results of testing small thick-walled cylinders under thermal shock conditions in the Heavy Section Steel Technology program, and of warm prestressing tests at the US Naval Research Laboratory. The analyses developed in these efforts have shown that operational pressure-temperature transients can be effectively modeled and applied to pressure vessel materials having appropriate mechanical properties simulating radiation embrittlement, and that the crack initiation/arrest conditions predicted will occur. Thus, the methodology has been developed for the analysis, and it shows continuing integrity of reactor vessels even in the presence of severe thermal shock transients. Thirdly, the technology of crack arrest has reached a level wherein standardization of test specimens and testing methods is now possible and, indeed, is underway. The detailed factors which affect arrest of running cracks, and how those factors must be evaluated to permit determination of a minimum toughness descriptive of crack arrest are better understood. Although this area of fracture mechanics technology still requires much validation testing, the currently available results are already being utilized in operating and accident analyses.