program - AIAS Summer School 2023

Prof. Nicola Bonora

University of Cassino and Southern Lazio, Cassino, Italy

The strength of materials in the presence of defects, better known as Fracture Mechanics, is a relatively recent science whose essential concepts were developed over a very short period and, almost immediately, became part of engineering practice. The lecture is organized in two parts: first, the essential concepts of the design-against-fracture from a historical perspective are presented, and successively, the most recent modeling developments, considering the different micromechanisms responsible for fracture in metals and alloys, are presented. Finally, several application examples are shown.

Prof. Chiara Bertolin

The lecture aims to provide an introductory overview of which are nowadays the typical applications of machine learning (ML) algorithms in heritage field to enhance the damage detection and predictive maintenance and preservation capabilities. It introduces what are the threats induced by climate change and/or by artificial climate control strategies on historic materials and structures. It progresses illustrating some of the damage functions usually adopted to evaluate climate-induced risks on materials and finally, it provides an overview of what are the novel data-driven approaches focused on the use of machine learning (ML) algorithms. ML algorithms can be used for several applications for example, for predicting the compressive strength of masonry or repair mortars, for highlighting possible damage scenarios in historical buildings caused by natural hazards, for determining the mechanical properties of materials, and superficial damages on monuments caused by weathering as a synergistic effect of mechanical, chemical, and biological decay. Finally, ML can be used as a reliable method to reconstruct corrupted and missing monitored data collected indoors in historical buildings, churches, and museums or to predict whether climatic fluctuations may have catastrophic effects on hygroscopic materials in terms of hygro-mechanical stress development.

Prof. Aurelio Somà

and

Dr. Davide Clerici

Generally, fracture is considered detrimental to the mechanical performance of materials and has received interests from scientists and engineers for centuries. Traditional chemistry-based method and micromechanics-based method have extensively been studied to enhance fracture resistance of materials, and both methods have difficulties to tackle current industrial challenges. Recently, bioinspired design method and machine-learning-based method have been advocated as promising methods against fracture. In this lecture, these two innovative design methods will be presented. The first part of this lecture will introduce strategies inspired by seedcoat in Nature, and the second part of this lecture will focus on how ML-based method goes beyond human’s engineering intuition on the design of lattice materials.