Conference Agenda

Tension members in major civil and building structures such as bridges, railways, and offshore structures serve as critical components for load transmission and structural stability assurance. However, long-term use, environmental exposure, and inadequate maintenance can lead to anomalies in tension force, resulting in an increasing trend of structural collapse incidents worldwide. This study aims to prevent such failures by developing an openable detachable split Yoke-type Elasto-Magnetic (E/M) sensor and verifying its performance for monitoring tension force in tension members. Previous Yoke-type E/M sensors were designed with only the primary excitation coil featuring an openable structure, while the secondary sensing coil had a non-openable structure, which necessitated on-site winding and limited practical field applicability. In this study, both the primary and secondary coils were designed with openable structures. COMSOL Multiphysics 3D simulations were performed to validate the design. The magnetic flux density of the sensor was measured using the Magnetic Fields module to verify magnetization characteristics, and based on these validation results, the sensor was fabricated through precision coil winding. To evaluate the sensor performance, induced voltage signals were acquired at 1-ton increments from 0 to 10 tons in both the COMSOL Multiphysics simulation environment and simultaneous physical experiments. The peak-to-peak voltage values were obtained by calculating the difference between the maximum and minimum values of the collected induced voltage signals. A regression analysis was then performed on the average peak-to-peak values to derive a quantitative tension force estimation formula. The derived formula was validated through comparison of simulation results with experimental results. Additionally, to enable future application to floating offshore wind turbine mooring lines, underwater environment simulation was performed based on electrical properties reflecting actual seawater conditions. Sensor performance variability according to external environment was analyzed, and practical field applicability was verified. This study advances sensor technology for structural health monitoring by proposing the design and fabrication of a fully openable, detachable split yoke-type E/M sensor and by experimentally and numerically establishing a quantitative relationship between induced voltage signals and tension force, thereby strengthening the academic foundation of non-destructive evaluation technology.

Strain gauges are widely used in Structural Health Monitoring (SHM) systems, due to their low cost and the extensive research that has been conducted on them. However, their highly localised measurements mean that many units are required. Measuring each strain gauge individually requires many measurement channels and extensive cabling, which can lead to a lot of manual labour and high costs. SHM systems are becoming increasingly relevant due to their potential cost and weight savings. At the same time, the amount of measurement data needs to be reduced.

The aim of this work is to develop a measuring system that can cover a large area with a small number of measuring points and, consequently, a small amount of measuring data. It must be demonstrated that the measuring system can distinguish between different types of load and damage. In order to demonstrate the benefits of reducing cables and data while ensuring that a single crack influences only a single strain gauge, a network of seven strain gauges has been chosen. Rather than measuring the strain of these strain gauges individually, they are interconnected according to the setup shown in Figure 1, forming a strain gauge network. Resistances are measured between the points (1-3), (1-4), (1-5) and (1-6).

First, an analytical calculation is conducted to determine the minimum number of measurement channels required for each level of SHM. It was found that four measurement channels are sufficient for load monitoring, as well as for detecting, localising and classifying a single crack in a network of seven strain gauges. While this may not seem like a significant reduction, measuring every strain gauge in larger networks would result in a substantial amount of cabling.

An experiment is conducted on a four-point bending beam. Highly sensitive resistance measurements were used to verify the analytical results. After tests are performed to verify the load monitoring results, a crack is induced in the middle of the beam, to demonstrate the network’s SHM capabilities.

Analytical calculations show that load monitoring and detection of different damage locations and sizes can be conducted using fewer measurement channels than strain gauges. Experimental results from a four-point bending test confirm these calculations. Future work will involve testing larger networks, as well as entire coatings, with the aim of eliminating the need for strain gauges and manual cabling entirely.

The Chain is discontinuous structures component widely used as critical components for power transmission and load bearing in various industrial facilities. However, local damage caused by repeated loading, friction, corrosion, and fatigue cracking can lead to catastrophic chain failure and unexpected equipment shutdown, resulting in severe safety risks. Accordingly, the objective of this research is to prevent chain failure-induced accidents by designing a Magnetic Flux Leakage (MFL) sensing system and experimentally verifying its capability to detect early-stage localized damage. To achieve this, an MFL sensor was designed to detect local damage non-destructively. A magnetization unit consisting of Neodymium permanent magnets and carbon steel yokes was applied to magnetically saturate the chain, and an array of 8-channel Hall sensors (positioned on the top and bottom) measured the leakage flux signals generated from the damaged areas. An experimental chain specimen of 3 m length with artificially machined defect depths between 1–4 mm is prepared, and leakage-flux measurement is repeated 400 times per damage level using a DAQ board for high-speed sequence acquisition. Because the acquired leakage sequence includes electrical noise and magnetization-induced baseline drift, a moving-average-based detrending method is applied to suppress non-diagnostic low-frequency components while preserving defect-related modulation. Fast Fourier Transform (FFT) analysis is performed to project the spatial magnetic leakage pattern into the frequency domain. Damage-governed spectral peaks that grow distinctly with defect severity emerge within constrained frequency bands, and these peaks are designated as core defect-signature frequency components. A band-pass filter retains only these core components, and Inverse FFT (IFFT) reconstructs them back to the spatial distance domain for direct visual confirmation of defect localization and severity-dependent amplitude variation. This study is expected to enhance the efficiency and reliability of chain maintenance and diagnosis through the design, fabrication, and verification of the MFL sensor. Consequently, this research contributes to the advancement of the Structural Health Monitoring (SHM) field.

The knowledge of the profiles of concrete water content is very important regarding the durability prediction of civil engineering reinforced concrete structures. To monitor changes in water content gradients in these structures, complementary ultrasonic, capacitive and electrical sensor systems were designed in the framework of the ANR-SCaNING research project. Eight common, useful depths were defined in order to obtain measurements in the same areas, in the cover concrete of the rebars and in the structure concrete core. The ultrasonic sensor system consists of eight pairs of transducers. The initial observable is the propagation velocity of the compression wave, while the observable monitored over time is the relative velocity variation [1]. The capacitive system consists of a double ladder with eight electrodes each. The observable is the dielectric constant which is the real part of the relative permittivity [2]. Finally, the resistive sensor consists of a multi-electrode ladder that can be interrogated in Wenner configuration or transmission configuration [3]. The observable is the electrical resistivity. These systems were embedded in a 100x100x30 cm concrete slab that was submitted to a three-month unidirectional drying in an oven at T=45°C. The gradients of the various observables are monitored over time and can be processed to obtain water content profiles.

The aim of this paper is to present the three different embedded sensor systems and to explain the processing approaches. The results will show the evolution of the gradients of the observables over time. Finally, the water content profiles obtained after three months of drying will be compared for the three sensor systems.

[1] Hariri R., Chaix J.-F., Shokouhi, P., Garnier, V., Saïdi-Muret, C., Durand, O., Abraham, O. (2024) Quantifcation of the Uncertainty in Ultrasonic Wave Speed in Concrete: Application to Temperature Monitoring with Embedded Transducers. Sensors 2024, 24, 5588. https://doi.org/10.3390/s24175588

[2] H. Ibrahim, G. Villain, N. Ranaivomanana, S. Palma Lopes, J.-P. Balayssac, T. Devie, X. Dérobert. (2024) Design and validation of a multi-electrode embedded capacitive sensor to monitor the electromagnetic properties in concrete structures, Measurement, Vol. 236, 2024, 115057. https://doi.org/10.1016/j.measurement.2024.115057

[3] Badr J., Fargier Y., Palma-Lopes S., Deby F., Balayssac JP., Delepine-Lesoille S., Cottineau LM., Villain G. (2019) Design and Validation of a Multi-Electrode Embedded Sensor to Monitor Resistivity Profiles over Depth in Concrete, Construction and Building Materials, Vol. 223, 30 October 2019, pp. 310-321. https://doi.org/10.1016/j.conbuildmat.2019.06.226