Local and Global Response Parameters in Seismic Risk Assessment of RC Frames Retrofitted by BRBs

This paper deals with the seismic performance and risk assessment of existing reinforced concrete (RC) buildings with limited ductility retrofitted by means of buckling restrained braces (BRBs). Two different approaches for evaluating the seismic vulnerability and risk before and after retrofit are introduced and analyzed. These approaches involve the use of different categories of engineering demand parameters (EDPs) for the system response assessment: global EDPs, that permit to obtain a synthetic description of the system behavior at a reduced computational cost, and local EDPs, more accurate in describing the response of the frame elements and of the BRBs, though more demanding from a computational point of view. A probabilistic methodology is first introduced that permits to evaluate and compare the vulnerabilities of the frame before and after the retrofit (Freddi et al. 2012). This methodology, similarly to other methodologies present in the literature (e.g., Hueste and Bai 2006, Ramamoorthy et al. 2006, Özel and Güneyisi 2012), is based on the development of system fragility curves before and after the retrofit. However, differently from the others, it employs local rather than global EDPs to monitor the structural seismic response. This approach allows to capture accurately the modifications of the frame components’ response induced by the added bracing system. Successively, the effects of the EDPs choice on the seismic risk assessment and risk-based design of the retrofit of existing RC frames with dissipative braces is investigated by considering and analyzing a specific case study. This consists of an existing RC frame with low ductility capacity retrofitted by inserting a system of BRBs with elasto-plastic behaviour. The braces are designed by applying a widespread method based on an equivalent nonlinear SDOF approximation (Dall'Asta et al. 2009) and by considering different values of the shear capacity of the bracing system. The importance of using local EDPs in the probabilistic evaluation of the retrofit effectiveness for the type of system analyzed in this paper is demonstrated by comparing both the fragility curves and the risk estimates under different hazard scenarios. It is shown that the use of global EDPs may result in a significant overestimation of the retrofit effectiveness in terms of both vulnerability and risk reduction. Consequently, if a risk-based design is carried out for the retrofit system, the dimension of the braces evaluated by using global EDPs for vulnerability assessment are significantly lower with respect to the corresponding dimensions obtained by using local EDPs. Larger differences are observed for the hazard scenarios with higher intensity.