Machine-learning model for the determination of macro-scale masonry properties based on a virtual laboratory at micro-scale

Abstract

Cutting-edge methods in the computational analysis of structures have been developed over the last decades. Such modern tools are helpful to assess the safety of existing buildings. Multi-scale techniques have been proposed to combine the accuracy of micro-modelling and the computational efficiency of macro-modelling. Machine-learning tools have been utilized successfully to train specific models by feeding big source data from different fields, e.g. autonomous driving, face recognition, etc. This research proposes a continuous nonlinear material law that can reproduce data from micro-scale analysis. The proposed method is based on a machine-learning tool that links the two scales of the analysis by training a macro-model smeared damage constitutive law through benchmark data from numerical tests derived from micro-models.