Modelling of historical masonry structures: comparison of different approaches through a case study

Abstract

In this paper, the applicability of different numerical techniques for the analysis of masonry structures is investigated, comparing the computed results with the experimental test data obtained on a full-scale masonry specimen. Three approaches are taken into account. Namely, the standard FEM modelling strategy, based on the concepts of homogenizated material and smeared cracking constitutive law, is used in the version implemented in the commercial code ABAQUS by HKS. The programme Visual CASTEM 2000 is used for the application of the finite element method with discontinuous elements, which are intended to simulate the presence of vertical and horizontal mortar joints. Finally, the UDEC software by the ITASCA Group is adopted for the Discrete Element modelling. An overview of such numerical methods, as well as a brief description of their specific theoretical aspects, is provided in order to allow easy comparison. Besides, ‘ready-to-use’ input parameter values, which are usually hard to decide, are suggested, basing both on extensive sensitivity analyses and experimental validation. Such values have proven to give reasonable results in the modelling of masonry structure under monotonously increasing loads, and for this reason can have practical usefulness.