Underwater Inspections to monitor the structural integrity are mandatory by regulations, due to casualties and the high financial costs associated with failure. This can be seen from the Alexander L Kielland platform capsizing incident in 1980 which happened due to the lack of inspection . Current visual inspection of underwater structures involves deploying divers or ROVs off vessels. Findings are then documented, and maintenance work is carried out to preserve its structural integrity. High costs of inspection deployment encourage asset owners to comply only to the bare minimum of regulatory standards – conducting inspections once every 1-5 years (according to major classification societies) , depending on the structure and its environment.
The predictive maintenance industry is expected to experience a huge growth from USD 4.0 billion in 2020 to USD 12.3 billion by 2025, at a Compound Annual Growth Rate (CAGR) of 25.2% . This trend indicates an increasing interest in continuous inspections to schedule pre-emptive maintenance before a breakdown – a job we believe is best done autonomously.
The concept of Autonomous Operation and Maintenance (AOM), refers to the concept of augmenting the capabilities of a human personnel to get the underwater inspection done more efficiently. Achieved using computers trained by machine learning to mimic human cognition, this makes processes smarter and faster, reducing the need for human intervention.
A floating solar farm, with multiple underwater elements to be inspected provides a strong use case for AOM. In a recent collaboration with G8, BeeX’s A.IKANBILIS was used to conduct inspections on the anchoring and mooring system of the solar farm. With AOM, asset owners can reduce their reliance on manual labour and improve the efficiency of data collection. In this case A.IKANBILIS was specifically inspecting the mooring ropes for marine growth and chafing along the cross-linkages, to ensure that they were holding well.
In general, inspection parameters and operations vary in accordance with:
These factors determine the complexity of inspections, in terms of maneuvering the AUV and collecting quality data.
Despite the low visibility conditions at the inspection site, A.IKANBILIS was able to hover close to the mooring rope and the anchors, while keeping a fixed offset distance with the use of its real-time imaging sonar.
Inspection efficiency and quality is increased, with A.IKANBILIS’ ability to return to a specific location during the next inspection to check for changes by referring to the recorded geo-references.