| description abstract | The seismic interaction of unanchored objects with a seismically qualified system or component in a nuclear power plant (NPP) can be detrimental to nuclear safety. In order to assess the seismic risk, an accurate quantification of the seismic response of unanchored components, such as spent-fuel dry storage casks, portable standby generators, electrical transformers, unreinforced masonry radiation shielding walls, etc., is required. The rocking response and toppling vulnerability of unanchored objects is not covered in nuclear standards, guidelines, and reports, except for an approximate method provided in ASCE/SEI 43-05 to estimate the maximum rocking angle in lieu of nonlinear time history analysis. In this study, the approximate method adopted by ASCE/SEI 43-05 is evaluated, first qualitatively and subsequently by comparing its predictions against results from a series of nonlinear time history analyses (NLTHA). The study concludes that the ASCE/SEI 43-05 method (which has also been adopted in FEMA P-58-1 and is expected to be adopted in the new revision of ASCE 4) provides highly unreliable, and in many cases unconservative, estimates of peak rocking rotation for a wide range of block geometries, under various levels of excitation. Because a reliable simplified method for estimating rocking rotations is not currently available, the paper recommends that rotation demands be obtained from nonlinear dynamic analysis. | |
