Virtual design of masonry triplet shear tests using digital image correlation

Abstract

Numerous studies highlight the diverse opportunities offered by multi-view image correlation techniques. For example, in masonry structures, these techniques enable the propagation of cracks to be quantified in joints and blocks. However, many parameters must be considered to transition from measurements to identifying physical parameters. Consequently, it is difficult to know a priori whether the planned experiment will enable for a precise identification of the model parameters to be calibrated. A good practice, detailed in this article, involves designing the experiment through virtual tests. The first step involves positioning the cameras in a virtual setup using Blender by combining near-field and far-field cameras for multiview purposes. Once a numerical model has been selected for nonlinear simulations, the second step involves examining the influence of geometric parameters to optimize the specimen response with respect to the quantities of interest. Last, virtual tests are conducted in Blender to simulate experiments under conditions close to reality. Based on the force data produced by the numerical simulations and the displacement fields measured from images generated during virtual tests, the proposed method enables for sensitivity analyses and uncertainty quantifications to ensure that the model parameters can be calibrated prior to the actual experimental campaign.