DNV GL has developed a solution that prevents the risk of offshore floating vessel mooring line failure going undetected using a machine learning algorithm.
DNV GL has introduced the mooring new solution designed to replace the physical sensors with a machine learning algorithm that accurately predicts line failure in real time.
The company said that its smart mooring solution addresses growing industry concern about the high frequency of mooring line failure, and a vessel’s subsequent loss of station.
Since the past two decades, more than 20 incidents have been reported across the world relating to the failure of permanent mooring systems on floating structures. In the most severe cases, vessels have drifted and risers have ruptured, causing extended field shutdown, and risk to life, property and the environment.
DNV GL said that the results shown by a numerical case study of a turret moored floating production, storage and offloading vessel (FPSO) with over 4,000 test cases have showed that its smart mooring solution is capable of accurately identify mooring line failure.
DNV GL – Oil & Gas Americas regional manager Frank Ketelaars said: “Our Smart Mooring solution can be deployed to predict a mooring system’s response to various operating conditions. It determines when a mooring line has failed, more accurately and cost-effectively than physical tension sensors currently used to detect anomalies.
“Conservatively, we estimate it is half the cost to implement our solution versus installing a mooring line tension monitoring system for a brownfield operation.”
The company said that the tension sensors are difficult to handle and costly to maintain, and are prone to failure within the first few years of installation according to the field experience.
The new smart mooring solution is capable of replacing failed sensors in brownfield offshore operations, or as a complete alternative to implementing sensor technology in greenfield offshore oil and gas developments.
The firm said: “DNV GL’s experts developed the Smart Mooring solution by training a machine learning model to interpret the response of a vessel’s mooring system to a set of environmental conditions and are then able to determine which mooring line has failed.”