Separation Method of Submarine Cable Fault Vibration Signals Based on EEMD-SVD-JADE

Abstract

Distributed optical fiber vibration sensing technology has been widely used to detect the fault of optical fiber composite submarine cable. Submarine cables are often damaged by anchor smashing, anchor dragging and wave scouring during operation. The vibration signals collected by optical fiber sensors usually show the mutual coupling of multiple fault vibration signals, so it is difficult to directly identify the types of fault vibrations. In order to separate vibration signals, a method based on EEMD-SV1D-JADE for fault signal separation is proposed. After de noising the coupled fault vibration signals, the vibration signals are firstly decomposed by EEMD to obtain several modal components. The SVD method is used to calculate the modal component matrix to obtain the number of vibration sources contained in the signals. Then, the dominant components are selected according to the number of vibration sources, which are input into the JADE algorithm as virtual channel signals for separation. Finally, the single source fault vibration signals are obtained. The validity and feasibility of the algorithm are verified by evaluating the consistency between the separated signals and the analog signals; By analyzing the separation effect of the algorithm under different signal combinations and different signal-to-noise ratios, it is proved that the algorithm has high stability; Field tests were carried out on the submarine cable operation site, and multiple groups of fault vibration signals were collected for separation and verification. The experimental results show that SVD can accurately estimate the number of vibration sources of submarine cable fault vibration signals, and single source fault vibration signals can be quickly separated through JADE algorithm. The separation results can help to realize the accurate identification of submarine cable fault types.