Offshore wind farms play a vital role for a sustainable, low-carbon future. Subsea power cables are an essential asset for the electrical transmission and distribution of power from such windfarms. The reliability of these cables determines the sustainability of the power supply and the economic viability of offshore wind farms.

Locating and replacing damaged sections of subsea cables can be very costly, as this example in the Western Isles shows. Subsea power cable faults are estimated to cause a loss of up £500,000 a year, due to lacking capabilities of early failure detection and lifetime prediction of these critical assets.

Heriot-Watt Smart System Group (SSG) tackles this issue with an innovative fusion prognostics approach for subsea power cable health management. We utilize multi-physical models, advanced sensing technology and data-driven techniques to conduct integrity assessment and predicting remaining useful life of this critical asset.

Multi-physical models simulate the processes leading to common subsea power cable failure and cable damage, i.e. abrasion and corrosion. Furthermore, simulation takes cable displacement and scouring into account, if subjected to various environmental conditions [1]. The Heriot-Watt team manufactured and implemented a novel, wide band low frequency sonar in order to collect cable integrity data. Using machine learning, we demonstrate the capabilities of the proposed approach to undertake a detailed and in-situ assessment of subsea cable integrity [2], [3]. Our approach provides a new assessment capability for the installation process, while allowing asset owners to better monitor and maintain power subsea cables.

[1] Flynn, D., Bailey, C., Rajaguru, P., Tang, W., & Yin, C. (2018). PHM of Subsea Cables. In M. G. Pecht, & M. Kang (Eds.), Prognostics and Health Management of Electronics: Fundamentals, Machine Learning, and the Internet of Things (1 ed., pp. 451-478). Wiley. https://doi.org/10.1002/9781119515326.ch16

[2] Tang, W., Flynn, D., Brown, K. E., Robu, V., & Zhao, X. (2020). The Application of Machine Learning and Low Frequency Sonar for Subsea Power Cable Integrity Evaluation. In OCEANS 2019 MTS/IEEE SEATTLE [8962840] IEEE. https://doi.org/10.23919/OCEANS40490.2019.8962840

[3] Tang, W., Flynn, D., Brown, K. E., Robu, V., & Zhao, X. (2020). The Design of a Fusion Prognostic Model and Health Management System for Subsea Power Cables. In OCEANS 2019 MTS/IEEE SEATTLE [8962816] IEEE. https://doi.org/10.23919/OCEANS40490.2019.8962816