A model for pushover analysis of confined masonry structures : implementation a...

• Autor(es): Marques, Rui Filipe Pedreira ; Lourenço, Paulo B.
• Data: 2013
• Identificador Persistente: http://hdl.handle.net/1822/26554
• Origem: RepositóriUM - Universidade do Minho
• Assunto(s): Seismic loading; Data mining; Shear strength; Wide-column model; Discrete element model; Quasi-static test
• Tipo de Documento: Artigo
• Idioma: Inglês

Descrição

Confined masonry (CM) is a typical building technique in Latin American countries. This technique, due to its simplicity of construction and similarity with traditional practices of reinforced concrete building, presents a potential of use in European regions with moderate-to-high seismicity. However, most of the procedures for seismic design in codes for Latin America are force-based, which appears to be inadequate due to the high dissipative response observed for CM. This paper presents a simplified numerical-analytical approach to model CM structures using pushover analysis, aiming to apply performance-based design procedures. First, a data mining process is performed on a database of experimental results collected from lateral tests on CM walls to adjust prediction models for the wall shear strength and to determine the input relevance through a sensitivity analysis. Then, an analytical model of CM structures for pushover analysis is proposed with basis on a wide-column approach that employs an adaptive shear load-displacement constitutive relation. The proposed method is compared with a discrete element model that represents explicitly the confinements-masonry interaction, against the experimental results obtained in a quasi-static test of a full-scale tridimensional CM structure. The accuracy of the predictions from both methods is very satisfactory, allowing to capture the base shear-displacement envelope and also the damage patterns of the structure, thus, demonstrating the ability of the methods to be used in performance-based seismic assessment and design of CM buildings.