<p>Offshore production facilities play a central role in the oil industry, considering the increasing demand for energy resources. Therefore, the necessity for monitoring these floating structures has increased, and the mooring system is a critical component for safety maintenance. The opportune identification of mooring lines failures can prevent environmental pollution, property losses, and&nbsp;other&nbsp;system failures. &nbsp;According to the accompanying paper in this conference, while traditional techniques rely on line tensions, new models propose monitoring the mooring system through the platform's positional data, extracting the information required to identify failures.&nbsp;This paper proposes a system to detect and classify the line failure&nbsp;based on the natural period's difference in the longitudinal axis and&nbsp;the lateral axis of the platform long drift oscillatory motion. The hypothesis is that when a line failure occurs, the natural period of the platform's oscillation increases and that this difference should indicate the malfunctioning of the mooring system. We built the proposed model using three modules: i) a natural period estimator, ii) a regressor, and iii) a classier.&nbsp;The estimator module (i) uses digital signal processing techniques, such as Welch's method, to extract the natural period from positional information (surge, sway, and yaw), considered input. The output is the correspondingly natural period. A 2-hour window, moving for 1 hour, is sampled from&nbsp;this time&nbsp;series, and&nbsp;this signal repeats three times.&nbsp;The regressor (ii)&nbsp;uses the GPS data to estimate the natural period's value assuming unbroken lines. We developed the regressor using a support vector machine with a radial kernel.&nbsp;The classier (iii) infers the line failure based on the difference between the natural period assuming&nbsp;unbroken&nbsp;lines (given by the regressor) and the natural period measured using the estimator. The data used to develop the system was provided by computational simulations using real environmental data:&nbsp;we used 18000 cases of dynamically simulated spread moored FPSO.&nbsp;Precisely, we used simulated data and artificial signals to test the estimator module&nbsp;(i); we used the 1-year simulation of 2003 data to train the regressor (ii) while the year of 2006 was the test set, and we split the 1-year simulation of 2006 into 80% train and 20% test set for the classier (iii).&nbsp;The system's metrics ensured an accuracy of 99% (+/- 1%) when the classifier received as input the natural period differences along the longitudinal and lateral axes; the average position of the platform, and the standard deviations of the 6 DoF, i.e., sway, surge, heave, roll, pitch, and yaw of the platform motion.&nbsp;&nbsp;The system classified the line failure with an accuracy of 75% (+/- 5%) using only the natural period information.&nbsp;The analysis of the results led to some conclusions.&nbsp;The first one is the validation of the initial hypothesis of the appearance of the natural period's difference when a line break and this difference's ability to indicate the failure. Furthermore, the proposed architecture worked precisely enough to guarantee the classification using the natural period.</p>

Fpso Mooring Line Integrity Supervising System Based on Motion Data and Natural Frequency Estimation

Paper Proceedings: https://asmedigitalcollection.asme.org/OMAE/OMAE2021/volume/85116

OMAE 2021

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.