Validation of numerical models for hospital building contents: rigid blocks and FEM models

The post-earthquake functionality of hospital buildings is an essential performance objective to achieve in a modern resilient community (Bruneau and Reinhorn, 2007). Such a performance may, however, be impaired due to the damage to non-structural components and building contents. Recent surveys carried out in the aftermath of major world-wide earthquakes (e.g. (Di Sarno et al., 2013, Jacques et al., 2014, Masi et al., 2014, among others) have shown that the overturning of cabinets, containing medical files with patient details, the failure of electronic panels is a typical non-structural component damage recorded after moderate-to-large earthquakes. Hazardous contaminants may also be released when medical cabinets and bookshelves overturn; hence there is a number of dangerous consequences caused by the lack of adequate seismic design. Comprehensive experimental and numerical studies were carried out in the last decade to investigate the seismic performance of a variety of furniture items, medical appliances and service utilities of typical hospital buildings and pharmacies, e.g. (Cosenza et al., 2014, Furukawa et al., 2013, Kuo et al., 2011). Numerical modeling of such components is still a key issue to be addressed. In this study, simplified finite element models of the tested components have also been implemented in software platforms to adequately simulate the dynamic properties of sample medical components. Different numerical modeling approaches are validated upon the outcomes of a comprehensive experimental campaign on hospital building contents carried out by shake table testing at the University of Naples Federico II, Italy. Finite element modeling approach is adopted to investigate the dynamic behavior of hospital cabinets in case they do not exhibit any rocking mechanism, i.e. pre-rocking behavior. The validation of a FEM model for the dynamic performance of cabinets is presented. Its ability to reproduce horizontal acceleration in the cabinets is also discussed. The developed numerical model gave a fairly good matching in terms of natural frequencies of the sample components. Nonlinear dynamic analyses are performed on the defined models. Recorded table acceleration are applied at the base of both the cabinets for the three different test groups. It is concluded that the defined model is able to recognize the occurrence of the rocking mechanism in the cabinets. Medical components, such as the tested cabinets, typically exhibits a rocking behavior as the seismic intensity increases. Thus, the dynamic behavior of rigid blocks is investigated for the post-rocking behavior of cabinets. Tested cabinets are modelled as equivalent rigid blocks and subjected to the experimental base accelerations. The ability to predict the occurrence of both rocking mechanism and overturning is verified. Given the good model fidelity, a preliminary study is performed aimed at the identification of the most efficient seismic intensity measure (IM) for rigid blocks and the influence of the geometric properties of rigid blocks on their dynamic performance