Conference Agenda

Structural health monitoring (SHM) using ultrasonic guided waves (UGWs) typically relies on dense sensor networks to enable reliable damage detection and localization, leading to increased system complexity and cost. To address these limitations, this study investigates the use of a frequency-steerable acoustic transducer (FSAT) as a compact alternative capable of directional wave excitation using a single actuator.
An experimental investigation is conducted on a conical steel structure representative of industrial geometries. A unidirectional FSAT is employed in a pitch-catch configuration together with a limited number of piezoelectric sensors distributed around the structure. The system is evaluated under pristine and damaged conditions, where a small blind hole is introduced as artificial defect. Guided wave responses are analyzed in both time and frequency domains over a broad frequency range (20–600~kHz), exploiting the FSAT’s frequency-dependent beam steering capability.
Damage detection is achieved through differential signal analysis and root-mean-square (RMS)-based damage indicators, which reveal clear sensitivity to the presence and location of the defect. Results demonstrate pronounced frequency-dependent behavior, with a distinct peak response at 350~kHz and significantly reduced sensitivity when the wavefield is steered away from the damage location, confirming directional selectivity.
The findings highlight the potential of FSAT-based SHM systems to reduce sensor network requirements while maintaining reliable damage detection and directional sensitivity in complex structural geometries. This work provides an experimental validation of FSAT technology in a realistic structural configuration and represents a step toward simplified and scalable SHM solutions.

This work investigates the capability of a permanent monitoring system to detect structural degradation before it compromises bridge operational requirements. The target structure is a multi-span steel truss railway bridge located in Italy, characterized by a complex geometry featuring Gerber connections and dual-deck configuration. Utilizing a high-fidelity finite element model of the entire structure, a comprehensive set of damage scenarios is simulated. The research evaluates the sensitivity of a monitoring setup consisting of inclinometers mounted on all three spans. A data processing pipeline is implemented to purge sensor signals from temperature-induced variations using multiple linear regression and principal component analysis. Damage detectability is then assessed by comparing damage-induced variations from ”virtual sensors” against the residual statistical variance (±2σ) derived from the real-world monitoring data of the healthy structure. The outcomes provide insights into the effectiveness of static monitoring systems for the identification of different damage conditions in complex truss bridges.

Climate change has become a critical challenge for maintenance and functionality of civil structures. Apart from global warming, climate change-induced frost periods can seriously affect dynamic behaviour of bridges. From a meteorological perspective, instability in the polar vortex and weakening of the jet stream are critical climatic phenomena that allow cold Arctic air to move toward mid-latitudes, leading to unexpected and severe freezing events. Under such circumstances, vibration features of bridges, especially modal frequencies, alter significantly. Generally, freezing temperatures cause sharp shifts in bridge modal frequencies due to sudden stiffening, particularly deck asphalts. Because these impacts can obscure true structural changes, produce false damage alarms, and reduce the reliability of long-term structural health monitoring (SHM), this paper proposes machine learning-aided data normalizers to mitigate climate change-induced frost conditions on modal frequencies of bridges during their SHM programs. Given the availability of temperature records, supervised data normalizers in terms of classifiers are trained by using both the modal frequencies and temperatures. In contrast, when temperature records are unavailable, unsupervised data normalizers in terms of reconstruction-based models are developed by the only bridge modal frequencies. Both types of data normalizers extract residuals between the measured and reconstructed modal frequencies, serving as normalized dynamic features free from frost effects. The proposed hybrid machine learning framework is validated using long-term monitoring data from concrete and steel bridges subjected to freezing conditions. Results show that this framework effectively eliminates frost-induced frequency jumps, enhancing the stability, accuracy, and climate resilience of long-term SHM programs.

The monitoring of defects using visual inspection is a complex and time-consuming task that is vital to the safety and reliability of infrastructures. This work proposes to tackle the automation of this task as a semi-supervised anomaly detection problem to overcome the limitations caused by the scarcity of defect samples. We choose to rely on normalizing Flows (NFs), a category of probabilistic models that map a source distribution to a target one, and that have recently proven to be very effective for anomaly detection. Most of these models have been tested on industrial inspection datasets and leverage neural networks pretrained on ImageNet to extract generic features before passing them to the NF to learn the distribution of healthy samples. In this paper, we investigate the use of a self-supervised feature extractor —a Masked Autoencoder (MAE)— to learn domain-specific features when the application domain differs significantly from ImageNet. To perform empirical comparaison, this encoder and two versions of the same architecture pretrained on Imagenet are used to train differents NFs. To evaluate the effectiveness of our approach, we then compute the resulting anomaly maps for each of them on concrete datasets.

Finite-lifetime structures such as aircrafts, bridges, and offshore wind turbines operate under highly variable environmental and operational conditions (EOCs). Studying the influence of these varying conditions on fatigue life remains a key focus in structural health monitoring (SHM) and life extension assessments of fatigue critical structures. Simulation-based or data-driven fatigue lifetime evaluation methods often struggle to represent the combined and nonlinear influence of multiple EOCs, such as wind, waves, and operational states. In particular when these still vary during the residual life of the asset. This motivates the development of visualization frameworks that can intuitively represent fatigue sensitivity across multi-dimensional EOC spaces, improving both interpretability and decision support for engineers and asset managers.

This study introduces a fatigue iso-life curves concept which is a visualization framework that maps the variations in fatigue life within selected environmental and operational parameter spaces. Each iso-life contour represents combinations of EOCs that yield the same predicted fatigue life, enabling a direct and interpretable assessment of how fatigue sensitivity evolves across operating regimes. The framework can be implemented in a wide range of applications, from aerospace to marine and civil structures, wherever fatigue governs long-term performance.

The concept is demonstrated through a detailed case study on offshore wind turbine (OWT) monopile foundations using SHM data from a 3 MW and a 9 MW turbine located in the Belgian North Sea. Six strain gauges installed on the transition piece recorded long-term structural response, which was combined with Supervisory Control and Data Acquisition (SCADA) including wind measurements and turbine operational parameters such as power, pitch, yaw, and rotational speed etc. Strain time series were processed into stress histograms, corrected for yaw orientation, and converted into fatigue damage using S-N curves and Palmgren-Miner’s rule.

A Random Forest (RF) regression model was trained on 12 months of strain and SCADA data, using wind speed, turbulence intensity and turbine operational state (operating/ parked) as input features and corresponding fatigue damage as outputs. The trained model was then used to perform a sensitivity analysis of OWT foundation fatigue life under varying wind speed distributions and turbine operational states.

The results were visualized as fatigue iso-life curves, showing as-measured operational regimes against as-designed wind conditions and turbine availability. These curves revealed distinct behavioural differences between the two turbines. For the 3 MW OWT, fatigue life consistently increased with greater turbine unavailability, when the turbine was parked, idling, or faulted, across all wind speeds, highlighting reduced loads during non-operational periods. In contrast, increased unavailability in 9 MW OWT led to longer fatigue life at lower wind speeds but caused a reduction at higher wind speeds, where parking the turbine introduced higher loads.

The proposed fatigue iso-life curve framework provides a powerful and generalizable means to represent fatigue sensitivity and uncertainty under variable environmental and operational conditions. Beyond the offshore wind sector, this visualization approach offers a scalable and interpretable tool for assessing fatigue performance in structures experiencing high environmental variability and evolving operating conditions.