Method and system for identifying external acoustic interference of optical fiber composite submarine cable

Abstract

The embodiment of the invention provides a method and system for identifying external acoustic interference of an optical fiber composite submarine cable, and belongs to the technical field of optical fiber composite submarine cables. The identification method comprises the following steps: acquiring Rayleigh back scattering light signal intensity of pulsed light; and acquiring a space-time sequence of the Rayleigh back scattering signal according to the signal intensity of the Rayleigh back scattering light. According to the method and the system for identifying the external acoustic interference of the optical fiber composite submarine cable, the Rayleigh back scattering light signal intensity and the space-time sequence of the pulsed light are monitored and obtained, the time-frequency two-dimensional image is obtained according to the space-time sequence, and the background noise is removed; meanwhile, the average power and the interference value of the Rayleigh back scattering signal are calculated, the occurrence condition of an interference event is judged, and the category of the interference event is identified by using a time-frequency two-dimensional image and a convolutional neural network; according to the method, background noise is effectively removed, and meanwhile, the recognition precision of interference events is also improved.

Description

technical field [0001] The invention relates to the technical field of optical fiber composite submarine cables, in particular to an identification method and an identification system for external acoustic interference of optical fiber composite submarine cables. Background technique [0002] Optical fiber composite submarine cable is the main equipment for providing power supply for offshore oil platforms and island residents, and for connecting offshore wind power to the land grid. However, the submarine environment for laying fiber optic composite submarine cables is complex and easily damaged by natural disasters such as ocean currents, earthquakes, tsunamis, and human factors such as ship anchoring and channel dredging. safe operation. The optical fiber is embedded in the fiber optic composite submarine cable, and the distributed acoustic wave information sensing technology is used to sense the influence of the sound wave near the fiber on the intensity of the Rayleigh...

AI Technical Summary

Problems solved by technology

In 2014, CNPC Langfang Pipeline R&D Center designed an identification algorithm based on wavelet transform for external force damage events of oil pipelines, using wavelet transform to decompose the original monitoring data into components of different frequency bands, characterized by the energy of different frequency bands, using the support Vector machine (SVM) recognizes and classifies external damage events such as climbing, cutting, and wall breaking, but wavelet transformation requires manual selection of wavelet functions and decomposition layers

In 2015, Tianjin University in my country proposed a method for identifying and classifying external force damage events that combines the Empirical Mode Decomposition (EMD) algorithm and the Radial Basis Function (RBF) neural network model. State function (IMF), using the energy ratio of each IMF as a feature, and then using the RBF model to identify and classify external force damage events such as climbing, knocking, cutting, shaking, etc., but EMD has discontinuous mode mixing, which will affect the identification precision

Subsequently, they used the zero-crossing rate and multi-scale combined entropy as features to improve the processing speed of the algorithm, and the average recognition time was only 0.32s. However, due to the need for manual feature extraction, the recognition and classification accuracy needs to be further improved

In 2017, they proposed an external force damage event recognition and classification algorithm based on mixed eigenvectors, which used frequency-domain bandwidth segmentation and kurtosis statistics as features, but due to the inability to completely eliminate all-phase digital filter bank sub-filters coupling, the correct recognition rate is still not high

[0005] The inventors of the present application found in the process of realizing the present invention that the above-mentioned solutions in the prior art have the defect that the recognition accuracy of external interference events is not high

Images