MS22.4 - Vibration-Based Assessment and SHM of Cultural Heritage Structure

The European historical built heritage largely consists of constructions made of masonry, such as palaces, towers, and churches. Ancient masonry is characterized by a complex mechanical response due to considerable material heterogeneity, a common presence of structural pathologies (either consisting of preexisting cracks or induced by aging and material degradation), as well as the occurrence of severe loading conditions such as those caused by differential soil settlements and seismic events. In this context, the lack of maintenance often plays an important role in aggravating and accelerating the development of structural pathologies. Assessing the structural integrity of historical masonry constructions during their lifetime is often a challenging task, yet of pivotal importance for management authorities who are responsible for (a) preservation of the cultural heritage, (b) life safety, and (c) scheduling of retrofit interventions. In this light, Structural Health Monitoring (SHM) approaches are becoming increasingly popular for the evaluation of the structural integrity of historical masonry constructions during operation. These techniques advocate the use of continuous monitoring of certain static/dynamic response parameters of structures to achieve a prompt identification of initiation/propagation of damage phenomena. Specifically, Operational Modal Analysis (OMA) methods are particularly well-suited for historical constructions given their non-destructive nature, global damage identification capabilities, and minimum intrusiveness upon the monitored assets. By processing their dynamic response under white noise ambient excitations, OMA methods allow characterizing the actual dynamic behavior of historical structures in terms of their modal features, i.e. vibration frequencies, damping ratios, and mode shapes. Despite their numerous benefits, the low sensitivity of OMA techniques to localized damage, as well as the difficulties in establishing robust correlations between changes in the modal features and the residual capacity of the structure, are often indicated in the literature as major limitations. This work is aimed at deepening into these aspects via an experimental program carried out on a full-scale masonry wall system with single opening, which was subjected to progressive damage. During the experimental activity, out-of-plane loading conditions of increasing magnitude were systematically applied to the wall specimen to induce damage initiation and progressive propagation. Ambient Vibration Tests (AVTs) were carried out on the specimen at each step to estimate the correlations between modal features and damage severity. The obtained results show clear variations in modal features consistent with the damage level suffered by the specimen. Interestingly, the variations in the mode shapes were found particularly sensitive to damage even at early stages of development. To further validate the experimental evidence, a Finite Element Model (FEM) of the specimen was developed and experimental results were replicated through non-linear modal analysis based on linear perturbation. The presented results demonstrate that FEM can mimic the damage-induced variations experimentally observed in the modal features of the specimen, providing a sound demonstration of the usefulness of modal features to identify slight-to-moderate damage.

The preservation of historic structures has become of primary importance in the civil engineering field due to their significant cultural value. However, destructive tests as well as invasive interventions should be limited as much as possible. In this contest, Finite Element model updating can be applied to optimize the numerical model by calibration of some unknown mechanical parameters. The solution of the inverse problem is obtained by minimizing the gap between corresponding experimental and numerical estimates of modal parameters. However, the application of Finite Element model updating is usually associated with large computational efforts, and the choice of a suitable surrogate model is fundamental for their reduction. Douglas-Reid method and Response Surface method are the most used due to their relatively simple quadratic formulations. While the approaches might appear quite similar, they have some peculiarities affecting their application and, in some cases, their performance. The present paper illustrates an application of Douglas-Reid method and Response Surface method to a historical tower, aimed at pointing out how the selected surrogate model affects model updating in terms of computational time and optimization results.

A new circle-shaped building is being built as an extension to the existing Viking Ship Museum in Oslo, part of the Museum of Cultural History, University of Oslo, Norway. The expansion will provide new exhibition areas, improving both the visitor experience and the physical conditions for the objects on display. Today, the museum has an average of 530 000 visitors per year and a floor area of 4000 m2. With the new extension, the area will be increased by 9300 m2, and the visitor capacity to more than 1 million per year. The museum houses the world’s largest collections of artefacts from the Viking era and holds the best-preserved Viking ships in the world. While smaller artefacts such as fabrics, jewelries and wooden objects of moderate size are moved off the premises and stored at a safe location, the three Viking ships and three sledges in the collection will be secured onsite during the construction work. The ships are too large to move off site, and the sledges are considered too fragile. Large and stiff steel rigs are mounted around the ships and vibration isolated at their current position. The sledges are moved to a specially constructed room located as far away from the construction site as possible, yet in the existing building, and conjointly placed on a vibration isolated stiff and heavy skid. Prior to the relocation of the sledges their support systems are improved to better cope with the dynamic and static loads introduced during the relocation and the following period of construction work. The mitigations are also designed with the long-term preservation strategy and future display situation in mind, with regards to both mechanical properties and aesthetics. Finite element models are developed as a tool for understanding the sledges’ static and dynamical behavior. The models are used to design effective mitigations applied to the sledges’ existing support systems. The models are established using highly detailed geometry 3D scans. They are dynamically calibrated using advanced vibration measurements identifying the sledges’ eigenfrequencies with corresponding mode shapes, and dynamic response to excitations. The models are also calibrated using load deformation measurements with controlled loads applied to the sledges’ support systems.