New Technology for the Mooring Failure Detection on Floaters

Abstract Over recent years, position-based mooring integrity monitoring systems have been extensively studied as they offer an efficient and cost-effective alternative to the conventional method using sensors attached to mooring components. However, there remains a gap between the overly simplistic classic watch circle methods and the excessively theoretical and resource-intensive machine learning solutions, suggesting the need for a different approach. The proposed solution has been developed internally based on the Company's experience in operating floating assets with different types of mooring, water depth and environment. This status-based method uses watch areas centered around damaged static coordinates in the horizontal plane instead of the intact position. The position and heading are corrected using wind and/or hawser tension measurements. The algorithm is complemented by a comprehensive validation methodology using synthetic line failures based on field measures and a complete calibration process to determine key parameters easily and accurately, either at design stage or for a retrofit. The methodology was successfully tested on various mooring configurations, including spread or single point mooring, catenary or semi-taut, deep or shallow water, and harsh or benign environments. Over 80,000 synthetic line failure scenarios were simulated based on a cumulated 10 years of measured data, and records from 18 actual mooring line failures or replacements were also analyzed. Generic and asset-specific settings or corrections were determined to detect all line failures, with zero false alarm and an average time-to-detect in line with the expected reliability of the failure detection and the emergency response, for a single- or double-line failure. And as the method is status-based, intermittent data availability is acceptable if the gaps are not excessively long, and any failure occurring during a gap will be detected afterwards. The detection system is currently being deployed on all Company floating assets, for both greenfield projects and as a replacement for legacy non-reliable traditional systems. This new method successively combines physics and statistical modelling within a simple and robust application, standing out against the recent machine learning solutions. Engineering and operational aspects are also connected thanks to a wholistic approach, ensuring that the detection is aligned with the required timeline of associated response. The calibration and validation process generates settings that meet the performance standards of a Safety and Environmental Critical Element and the consequent response requirements.