Digital twin for virtual sensing of ferry quays via a Gaussian Process Latent Force Model

2025-06-17

AI Summary

Ferry terminals rapidly deteriorate under ship impact and harsh marine environments, yet structural health monitoring via vibration-based methods is hindered by practical constraints on sensor deployment at critical locations. To address this, we propose a physics-informed machine learning framework that couples a reduced-order physical model with a Gaussian Process Latent Force Model (GPLFM) to jointly invert unknown impact characteristics—including location, direction, and magnitude—as well as their modal contributions, while accommodating time-varying boundary conditions. We further introduce a novel backward sequential sensor placement strategy to systematically evaluate the sensitivity of sensor type, sampling rate, and damping ratio. Experimental validation demonstrates that the digital twin enables high-accuracy acceleration response estimation at most measurement points, with controllable errors in localized impact zones. Notably, the method retains strong robustness and generalizability even under linear time-invariant simplifying assumptions.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Intelligent Robots: State Estimation

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Ferry quays experience rapid deterioration due to their exposure to harsh maritime environments and ferry impacts. Vibration-based structural health monitoring offers a valuable approach to assessing structural integrity and understanding the structural implications of these impacts. However, practical limitations often restrict sensor placement at critical locations. Consequently, virtual sensing techniques become essential for establishing a Digital Twin and estimating the structural response. This study investigates the application of the Gaussian Process Latent Force Model (GPLFM) for virtual sensing on the Magerholm ferry quay, combining in-operation experimental data collected during a ferry impact with a detailed physics-based model. The proposed Physics-Encoded Machine Learning model integrates a reduced-order structural model with a data-driven GPLFM representing the unknown impact forces via their modal contributions. Significant challenges are addressed for the development of the Digital Twin of the ferry quay, including unknown impact characteristics (location, direction, intensity), time-varying boundary conditions, and sparse sensor configurations. Results show that the GPLFM provides accurate acceleration response estimates at most locations, even under simplifying modeling assumptions such as linear time-invariant behavior during the impact phase. Lower accuracy was observed at locations in the impact zone. A numerical study was conducted to explore an optimal real-world sensor placement strategy using a Backward Sequential Sensor Placement approach. Sensitivity analyses were conducted to examine the influence of sensor types, sampling frequencies, and incorrectly assumed damping ratios. The results suggest that the GP latent forces can help accommodate modeling and measurement uncertainties, maintaining acceptable estimation accuracy across scenarios.

Problem

• Estimating ferry quay structural response with sparse sensors
• Modeling unknown impact forces using Physics-Encoded Machine Learning
• Addressing time-varying conditions and sensor limitations for Digital Twins

Innovation

• Gaussian Process Latent Force Model for virtual sensing
• Physics-Encoded Machine Learning integrates data and models
• Backward Sequential Sensor Placement optimizes monitoring