Abstract
The load-carrying capability of marine steel structures often suffers degradation and at times structural failure during their service life due to the corrosive sea environment and the extreme wave loads which they experience. Structural integrity is a high priority during asset design and operation to preserve human safety, environment protection, and to maintain operations.
Structural Health Monitoring (SHM), a process of deriving the health status and predicting structural damage via sensor-based measurements, data trending and analysis, has been commonly implemented in marine and offshore assets for decades. There are mature commercial SHM products on the market and available international and industrial regulations, guidance and classification rules (such as IMO MSC/Circ. 646, ABS Guide for Hull Condition Monitoring Systems, etc.). With the emergence of the Structural Digital Twin (SDT) concept in the marine industry in the recent years, the application of structural sensors on marine structures and the integration between sensor-based SHM techniques and structural digital twin approaches have attracted more attention. As a digital representation of the physical asset, the structural digital twin typically involves multiple-scale, multiple-physics, and data driven models and simulations. Leveraging the high fidelity full-scale SHM data, the structural digital twin is capable to provide more accurate assessments, predictions, and insights on structural integrity conditions and support the decision-making process regarding asset operation, structural inspection, repairs, and condition-based asset integrity management.
This paper reviews the mature and commonly employed SHM sensor techniques (such as electric resistance strain gauge, fiber Bragg grating (FBG) strain gauge, accelerometer, pressure transducer, etc.) and discusses their implementation on marine assets and integration with structural digital twin. The paper elaborates to provide some general guidance on the selection of the structural sensors and their installation, as well as the proper sensor specification, in the context of a structural digital twin implementation. The paper also discusses the required engineering mindsets and skillsets for developing and deploying sensor plans, and for implementing a sensor-based structural digital twin framework. The paper concludes that a successful SHM plan and structural digital twin implementation rely on sensor planning, data acquisition, data processing, analytic models and a proper integration of the data and data analytics. The overall planning and implementation should be governed by the suitability for the business purpose and implementation expectation.
