Stick-IT: A simplified model for rapid estimation of IDR and PFA for existing low-rise symmetric infilled RC building typologies

Abstract

The prediction of seismic performances of buildings in terms of engineering demand parameters EDP, such as interstorey drift ratios IDR and peak floor accelerations PFA, represents a fundamental step towards the assessment of potential direct economic losses.This paper proposes a simplified model for the rapid assessment of EDPs in infilled moment resisting frames subjected to seismic loadings suitable for large scale assessment studies. The proposed model, named Stick-I (Stick model for Infilled frames), is a multi-degree of freedom MDOF system consisting of a series of lumped masses connected by means of nonlinear shear link elements. The shear link behavior is suitably calibrated adopting a multi-objective Genetic Algorithm procedure that employs the results of nonlinear cyclic pushover analyses performed on refined nonlinear FEM. Based on the regression study performed on a suitably generated Stick-I model database, a more general Stick-IT model, representative of RC Infilled frame Typologies of assigned storey number and dimensions in plan, is introduced. Simplified formulas for the definition of Stick-IT model are proposed depending on low-level information data that can be easily retrieved from rapid on-site surveys or remote sensing techniques.Both Stick-I and Stick-IT models are validated comparing the results of NRHA performed on refined FEMs with the results obtained adopting the simplified models and good agreement in terms of maximum EDP values and distribution along the building height is obtained.The typological Stick-IT model, which is defined as a function of few geometric and mechanical parameters, can be easily applied for simplified evaluation of expected response for building typologies. Hence, it can also be usefully employed for assessment of direct losses at the large scale, with a significant reduction in computational burden with respect to more refined methods.