In this paper the actual dynamic behavior of the civic Clock tower of Rotella, a little village in central Italy heavily damaged by the recent 2016 seismic sequence, is thoroughly investigated by means of a detailed numerical model built and calibrated using the experimental modal properties obtained through Ambient Vibration Tests. The goal is to update the uncertain parameters of two behavioral material models applied to the Finite Element Model (elastic moduli, mass densities, constraints, and boundary conditions) to minimize the discrepancy between experimental and numerical dynamic features. A sensitivity analysis was performed with the definition of a metamodel to reduce the computational strain and try to define the necessary parameters to use for the calibration process. Due to the high nonlinear dependency of the objective function of this optimization problem on the parameters, and the likely possibility to get trapped in local minima, a machine learning approach was meant. A fully automated Finite Element Model updating procedure based on genetic algorithms and global optimization is used, leading to tower uncertain parameters identification. The results allowed to create a reference numerical replica of the structure in its actual health state and to assess its dynamic performances allowing better control over their future evolution.
This chapter introduces the subject of structural health monitoring (SHM) and discusses its important role in the post-earthquake assessment of damaged built heritage. The first part of the chapter describes the SHM operational workflow and provides practical advices for each of the different stages composing the procedure, while the second part exemplifies a real case-study application. The investigated structure is a cultural heritage building severely damaged by the Central Italy seismic sequence of 2016–2017. The steps leading to the design of the SHM system architecture installed in the building and to the definition of the most appropriate sensor network topology are presented, along with a preliminary insight into selected results that will act as comparative baseline information throughout the post-earthquake recovery phase of the monitored structure until completion of the restoration process.
COMBINING OPERATIONAL MODAL ANALYSIS AND GENETIC ALGORITHMS TO UNDERSTAND THE ACTUAL STRUCTURAL BEHAVIOR OF HISTORICAL CONSTRUCTIONS - 8th International Conference on Computational Methods in Structural Dynamics and Earthquake Engineering - 28-30 June 2021, Streamed from Athens, Greece
1 January 2024 in International Journal of Masonry Research and Innovation
The detection and anticipation of the gradual degradation of structures, whether caused by natural material deterioration or unforeseen vibrations such as those triggered by earthquakes or fluctuations in climate, are paramount in ensuring structural integrity. Nonetheless, the task of recognizing structural damage by scrutinizing changes in modal parameters, while simultaneously mitigating external factors such as environmental influences, poses significant challenges and time constraints. A prompt recognition of a structure's evolving behavior is crucial for ensuring human lives safety and implementing precise and timely interventions, whose rapid application in critical situations can prevent prolonged disruptions in a structure's functionality and expedite recovery efforts, where feasible. In this study, the effect of availment of methodologies that directly analyze data obtained from a continuous monitoring system deployed on a civic clock tower in Central Italy. This approach reduces computational burden and provides nearly instantaneous feedback on shifts in dynamic behavior. These considerations are pivotal for establishing sustainable monitoring practices that are both cost-effective and time-efficient