Sliding fragility of unrestrained equipment in critical facilities

ABSTRACT

Through the years, seismic design of buildings has been well developed and is continually updated and improved. Yet, nonstructural components housed in buildings are rarely designed with the same degree of consideration as buildings. As a result, buildings that remain structurally sound after a strong earthquake often lose their operational capabilities due to damage to their nonstructural components, such as piping systems, communication equipment and other types of components. The recent 1994 Nothridge, 1995 Kobe and 1999 Turkey and Taiwan earthquakes further demonstrate the importance of controlling damage to nonstructural components, particularly in critical facilities, such as hospitals, in order to ensure their funcionality during and after a major earthquake. Earthquake vulnerability of nonstructural components is usually reduced by fastening or bracing individual objects. However, there are some nonstructural components in buildings which often cannot be restrained for protection from earthquake shaking. The response of these objects will consist of sliding, rocking or jumping. Understanding these response types will allow estimation of vulnerability to earthquake damage and will assist in the design of appropiate mitigation measures. This research concentrates on experimental and analytical studies of the sliding response of freestanding rigid objects subjected to base excitation. Analytical and experimental techniques are combined to allow determination of fragility curves for free-standing rigid equipment under seismic excitations for further improvement of seismic mitigation measures. A discrete system model, an analytical model for two-dimensional sliding under two-dimensional excitation, is developed and analyzed for specific base motions. Shaking table testing with a range of excitations and systems parameters is used to define stability bounds for pure sliding motion. A comparison of the analytical and experimental results is then performed to further verify the validity of the analytical model. Discrepancies in the model assumptions and future improvements of the nonstructural model are also discussed in this report (AU)