24 L. Sibille et al. including their modal parameters, such as natural frequencies and mode shapes. These modal parameters are essential indicators of a structure’s health and are often used to detect changes in stiffness, mass distribution, or boundary conditions [7, 8]. Therefore, the assessment of the health of the structure requires filtering out environmental effects, particularly if they are expected to have a significant influence [9]. SHM techniques should consider these environmental influences [10] and apply appropriate filtering to ensure that variations in modal parameters reflect the structural condition rather than external environmental factors. However, the methodologies employed for system identification and environmental effect filtering must respect certain conditions. If modal parameters are found to be dependent on environmental effects, such as the periodic nature of tidal variations, or operational loads, such as the adjusting position maneuvers of linkspans of ferry quays, the assumptions for performing common System Identification (SI) algorithms, such as the Stochastic Subspace Identification (SSI) method, may no longer be valid. In such cases, it is better to use SI algorithms that do not require respecting the white noise condition, allowing for a more accurate identification of modal parameters. To explore this challenge, the Magerholm ferry quay was instrumented with an SHM system in 2023, transforming it into the Magerholm Research Quay (MQR). The monitoring system, implemented in collaboration with Møre og Romsdal County and the Norwegian University of Science and Technology (NTNU), consists of 22 accelerometers, 2 GPS antennas for synchronization, and 1 thermometer for evaluating thermal effects. The system aims to provide insights into how ferry quays are influenced by EOVs and tools for hazard alarming and damage detection. This paper is a preliminary study that aims to evaluate the correlation between tidal levels and the modal parameters, particularly the first natural frequency, of linkspans for the first time. It gives an overview of the monitoring system that has been set up at the Magerholm Research Quay. It describes its design, sensor layout, and data acquisition system. Additionally, it presents a detailed analysis of time histories of the acceleration responses recorded on the linkspan. An Automated Operational Modal Analysis (AOMA) algorithm based on the SSI method was applied to 20-second-long recordings collected daily over a six-month period. Next, the correlation between tidal levels and the first identified natural frequency of the structure was investigated by performing Least Squares Regression. Additionally, Principal Component Analysis (PCA) was performed to explore the temporal fluctuations of the principal components and separate the tidal effects from the natural frequency. The Magerholm Research Quay The MRQ is a collaborative project between Møre og Romsdal County and NTNU. The project aims to use real-time data collected from the structure to investigate quays’ structural behavior and optimize maintenance practices. Situated along Route 60, the Magerholm ferry quay serves as a critical link between Magerholm and Sykkylven. This quay transports around 950,000 vehicles annually, making it the fifth most utilized quay in Norway for vehicle traffic. The quay, along with its main components, is illustrated in Fig 1. The quay consists of two primary structures: a reinforced concrete pier and a steel linkspan. Originally constructed in 1997, it underwent a major renovation in 2015 to handle larger ferries, including a 25-meter extension of the concrete pier and modifications to the linkspan. The pier now spans 110 meters and is supported by end-bearing circular steel piles filled with C45 concrete, used for both the deck and the piles. Thirteen fenders are distributed across the quay to absorb ferry-induced impacts. In response to the upgrade to electric ferries, a charging station has been installed on the pier. A vacuum anchoring system provides stability, minimizing sway during loading and unloading operations. The steel linkspan, measuring approximately 15 meters in length and 9 meters in width, facilitates vehicle and passenger transfers. It incorporates longitudinal beams made from S355 steel and cross beams of S235 steel, with a front beam made of Hardox400 steel to resist the powerful ferry impacts. The linkspan is supported vertically by a concrete abutment and two lifting towers. The lifting towers, operated by a ferry operator, adjust the linkspan’s position to account for water-level changes such as tides and waves. Four conical fenders, positioned between the linkspan and the concrete foundation, absorb longitudinal movements in the surge direction. The monitoring system comprises 22 accelerometers, 2 GPS antennas, and 1 thermometer. The SHM system is illustrated in Fig 2. On the pier, 5 submersible uni-axial (Dytran 3211B2) sensors, 3 uni-axial (DY3055D3) sensors, and 4 tri-axial (DY3233A) sensors from Dytran, with sensitivities of 100 mV/g, 500 mV/g, and 1000 mV/g, respectively, are installed. Additionally, 10 submersible uni-axial (Dytran 3211B2) sensors are positioned on the linkspan, where they are exposed to waves during high tide and are partially submerged during storm surges. The sensors on the pier are labelled with ”P”, while those on the linkspan are labelled with ”L”. Uni-axial sensors are identified by ”T” for transversal (sway direction), ”L” for longitudinal (surge), and ”V” for vertical (heave) orientations, while tri-axial sensors are marked with ”A” as the final letter. All the sensors deployed are Integrated Electronic
RkJQdWJsaXNoZXIy MTMzNzEzMQ==