Genetic algorithms applied to nonlinear analysis for the identification of masonry structures collapse mechanisms

We investigate the promising capabilities of genetic algorithms (GA) for identifying collapse mechanisms in masonry buildings. This work deepens the knowledge of seismic behavior through the case study of the Padula palace, which is in the old town of Acri, in the province of Cosenza (Italy). The study of the structural behavior was performed by numerical modeling, developed by FaTaNext software, based on macroelements, and then by a nonlinear static analysis (pushover) by using Abaqus software. The obtained results were used in a second step for the determination and identification of the collapse mechanisms with the greatest probability of activation using genetic algorithms with Grasshopper. The validity of both methodologies is found in the satisfaction of the requirements set by the standards and in the actual ability to define plausible global behavior. An innovative procedure was proposed for the seismic evaluation, studying the possible modes of structural collapse with the aid of GA. This parametric method allows an effective collapse identification for the elevated structures that have greater fragility, allowing design attention to be focused on the elements characterized by a greater probability of collapse.