Holistic Subsea Cable Fault Detection & Condition Monitoring System

Power generation from offshore wind is subject to harsher environments than onshore wind, which can cause greater challenges for the reliability of critical infrastructure. One asset that is fundamental to the economic success of an offshore wind farm is the subsea cable infrastructure. This includes intra-array and export cables, failure of which will lead to significant loss of revenues through failure to export generated power, as well as penalties imposed by regulators. Since 2013, 8 TWh of electricity generated offshore has been lost due to subsea cable faults resulting in £\>1bn of losses for operators. In the offshore wind industry, 80% of the total cost of insurance claims are attributed to cable failures, while subsea cable expenditure only accounts for 9% of the capital cost of a new windfarm. Therefore, better cables or enhanced failure prognostics and monitoring systems are required to reduce the number of detectable cable failures and minimise the lost generation in the event of a fault. At present, no existing system is capable of accurately identifying, locating and offering mitigation in the event that a fault is detected.

SmartCable will bring a radical solution to subsea cable condition monitoring, with Proserv integrating numerous state-of-the-art monitoring systems and the data they generate to provide operators with, for the first time, a complete picture of subsea cable heath for predictive maintenance with cost savings associated with reduced failures, and optimised repair scheduling.

This will enable for the first time a cable monitoring system that can be deployed on subsea cabling for new and existing offshore wind farms. The SmartCable system, to be developed by Proserv will act as a decision-enabling platform to take the wealth of data -- temperature, electro-mechanical, partial discharge and acoustic data from the subsea cable fibre optic and main cores -- and correlate it centrally, developing a new common data language and applying bespoke machine learning algorithms to provide actionable intelligence to owner-operators on cable condition, and detect an operational fault at the earliest possible stage, long before a failure will take place. This will enable effective prediction of cable failure, allowing for contingency actions to be taken to reduce the risk of failure and plan for repairs.

Analysis of the performance metrics and system data during a demonstration at Levenmouth Demonstration Turbine(LDT)will allow the long-term robustness and stability of the system to be validated in an offshore environment and optimised prior to commercialisation.