Corrado Chisari1, Lorenzo Macorini1, Bassam A. Izzuddin1, and Claudio Amadio2
1) Department of Civil and Environmental Engineering, Imperial College London, London SW7 2AZ,
United Kingdom
e-mail: {c.chisari12, l.macorini, b.izzuddin}@imperial.ac.uk
2) Department of Engineering and Architecture, University of Trieste, 34127 Trieste, Italy
amadio@units.it
Keywords: brick-masonry; in-situ flat-jack test; mesoscale masonry model; inverse analysis; numerical optimisation; surrogate descriptions.
Abstract. Detailed mesoscale models enable realistic response predictions of masonry structures subjected to different loading conditions. The accuracy of the numerical predictions strongly depends upon the calibration of the model material parameters, which is usually conducted at the level of masonry constituents. However, especially for existing structures testing of individual components can be difficult or unreliable. In this work, an innovative approach for the calibration of a mesoscale masonry representation is proposed. It is based on the inverse analysis of the results of physical in situ tests performed using an innovative setup with flat-jacks. The post-processing inverse procedure comprises (i) metamodeling as a replacement of expensive nonlinear simulations, (ii) sensitivity analysis to reduce the parameters to identify to those which effectively control the recorded response, and (iii) optimisation by means of Genetic Algorithms to find the best fitting model parameter set. The potential of the proposed calibration procedure is shown considering the response of masonry components tested in laboratory following the proposed in-situ test.