SearcharxivSearch

arXiv subjects

S. Caddemi

Publications and source records attributed to S. Caddemi.

2 recordsLinked to original sources

A Discrete Macro-Element Method (DMEM) for the nonlinear structural assessment of masonry arches

The structural response of masonry arches is strongly dominated by the arch geometry, the stone block dimensions and the interaction with backfill material or surrounding walls. Due to their intrinsic discontinuous nature, the nonlinear structural response of these key historical structures can be efficiently modelled in the context of discrete element approaches. Smeared crack finite elements models, based on the assumption of homogenised media and spread plasticity, fail to rigorously predict the actual collapse behaviour of such structures, that are generally governed by rocking and sliding mechanisms along mortar joints between stone blocks. In this paper a new Discrete Macro-Element Method (DMEM) for predicting the nonlinear structural behaviour of masonry arches is proposed. The method is based on a macro-element discretization in which each plane element interacts with the adjacent elements through zero-thickness interfaces and whose internal deformability is related to a single degree of freedom only. Both experimental and numerical validations show the capability of the proposed approach to be applied for the prediction of the non-linear response of masonry arch structures under different loading conditions.

cs.CE

On profile reconstruction of Euler-Bernoulli beams by means of an energy based genetic algorithm

This paper studies the inverse problem related to the identification of the flexural stiffness of an Euler Bernoulli beam in order to reconstruct its profile starting from available response data. The proposed identification procedure makes use of energy measurements and is based on the application of a closed form solution for the static displacements of multi-stepped beams. This solution allows to easily calculate the energy related to beams modeled with arbitrary multi-step shapes subjected to a transversal roving force, and to compare it with the correspondent data obtained through direct measurements on real beams. The optimal solution which minimizes the difference between measured and calculated data is then sought by means of genetic algorithms. In the paper several different stepped beams are investigated showing that the proposed procedure allows in many cases to identify the exact beam profile. However it is shown that in some other cases different multi-step profiles may correspond to very similar static responses, and therefore to comparable minima in the optimization problem, thus complicating the profile identification problem.

cs.CE