Abstract

The assessment and the conservation of historical masonry structures are very challenging issues. According to the actual methodology, all the phases of the entire process of assessment require efforts and reciprocal comparison in order to understand reliably the structural behaviour and to design effective interventions. This thesis goes through such phases (anamnesis, diagnosis and treatment), introducing some innovations in each step and connecting the experimental experiences to models with the support of some real cases. Three techniques are developed about the knowledge phase, namely sonic test, flat jacks and dynamic identification. Deeper studies have been dedicated to vault systems by means of an extended experimental campaign with five full scale vaults tests and a reverse analysis to better understand the behaviour of structures and, at the same time, the limits of models. Sometime this comparison suggests a lack between model parameters and physical meaning due to modelling approaches (mesh, element type) and parameters (material properties and constitutive laws) and this gap may be fulfil by both local and global tests. From the experimental point of view this work presents a wide range of tests from the local to the global behaviour and varying among non-destructive, minor destructive and destructive tests. On the other hand for models, both linear and non-linear approaches have been adopted looking as well to local and global phenomena. Finally, about the deepest analysis on vaults even the scale of modelling was evaluated with the comparison between macro and meso-scale modelling. In this framework some proposal in kinematic analysis of strengthened vaults were provided. The work carried out allows therefore to compare traditional and more used tools for structural assessment purposes with real and measured experiences helping to validate the current methodology in the safety evaluation of existing buildings.