Sensing Technologies and Artificial Intelligence for Subsea Power Cable Asset Management

In this paper, we present two novel sensor systems to support the integrity analysis and asset management of subsea power cables. Firstly, we provide an example of a customized in-situ monitoring collar for subsea power cable monitoring, representing a first in full displacement monitoring of subsea power cables. Secondly, we provide results from an advanced Low Frequency (LF) sonar system, demonstrating the technology’s ability to differentiate different power cable types and varying levels of degradation. Results from laboratory experiments verify the ability of the monitoring collar to wirelessly monitor cable dynamics and an accuracy of 95% from LF Sonar analysis utilizing the state-of-the-art deep learning method (Convolution Neural Network) for detecting different level of cable damage. Our results provide a new ability to perform in-situ cross sectional analysis of subsea cables, and our monitoring collar concept can provide new information into real-time cable dynamics.