Developing an AI-based modeling tool to support decision-making in iceberg tow management for the protection of subsea infrastructures

Icebergs pose serious risks to subsea infrastructures operating in the Arctic and Atlantic regions. They can damage pipelines, cables, and foundations by scraping the seabed or directly colliding with offshore structures like floating systems and platforms. To prevent such impacts, icebergs are towed away from the facilities using specialized vessels and equipment. However, this process is complex, costly, and uncertain.
This project aims to develop an AI-based tool to support decision-making in iceberg tow management. The tool uses machine learning (ML) technology to predict the iceberg draft (a submerged portion of the iceberg) using the above-water features of the iceberg, which are captured by field measurements and remote sensing technologies. The tool predicts the subgouge soil deformations and reaction forces caused by the iceberg keel as it moves along the seabed. These predictions assist in optimizing the towing strategy and minimizing the risk of damage to subsea infrastructures.

Faculty Supervisor:

Hodjat Shiri

Student:

Partner:

Springboard Atlantic Inc.

Discipline:

Engineering

Sector:

Artificial Intelligence; Technology; Oil and Gas

University:

Memorial University of Newfoundland

Program: