Research On Vibration Signal Analysis Method Of Optical Fiber Composite Submarine Cable Based On ESMD And SVM

The composite submarine cable of the optical fiber,as the important transmitting approach to the electric power and communication for the offshore power grid,plays an important role in the realization of the interconnection of the regional power grid.Because the submarine cables are extremely vulnerable to the damage under the influence of complex marine environment and construction activities,the monitoring of the cable vibration caused by the submarine cable failure is necessary.This paper uses ANSYS finite element analysis software to study the vibration characteristics of the submarine cable under anchoring conditions and realizes the experimental simulation and identification analysis of the submarine cable fault vibration signal.Firstly,this paper establishes and solves the finite element model of the submarine cable operated by the anchor smashing within the ANSYS framework by means of setting the contacted pair of the submarine cable and the anchor,facilitating the loads and boundary constraints,based on the contact theory and the on-spot anchor smashing situations of the submarine cable.Secondly,the paper analyzes and summarize the several typical types of submarine cable faults by reviewing the literature at home and abroad from which the vibrating signal data about the scouring and friction effect originated from the oceanic wave are extracted.Thirdly,based on the nonlinear and unsteady features of the vibrating signal created by the submarine cable faults,the paper decomposes the signal into the finite Intrinsic Mode Functions(IMF),through utilizing the algorithm of the Extreme-Point Symmetric Mode Decomposition(ESMD)and extract respectively from IMF the three characteristics,that is,the energy,the sheath degree and the entropy of the enveloping spectrum.Furthermore,the paper constructs the experimental system simulating the cable's vibration created by the faults of the submarine cable.three typical submarine cable fault vibration signals,anchoring,scouring,and friction,are applied to the experimental fiber to observe the influence on the polarization state of the fiber.The polarization state change signal is collected as the experimental data,and the ESMD algorithm is used to decompose the signal.Extract features to form feature vectors as input SVM samples.Eventually,the extracted feature vectors are imported into the SVM optimized by particle swarm optimization and cross-validation,and the identification of the submarine cable fault vibration signal is realized.By means of finite element modeling and simulation,this paper obtains the characteristics of the vibration signal of the submarine cable under the action of anchoring,and combines the pole symmetry modal decomposition algorithm and the support vector machine algorithm to realize the classification and diagnosis of the submarine cable fault vibration signal.The classification effect is about 97%,which verifies the feasibility of this method.