A Distributed Optical Fiber Vibration Signal Feature Extraction and Recognition Method

Abstract

The invention discloses a distributed optical fiber vibration signal feature extraction and identification method, which belongs to the field of optical fiber sensing signal processing. First, the space-time matrix signal of a vibration source is collected, the spatial column signal is extracted, and short-term signal units are divided to construct optical cable vibration event data. Then build, train and tune an improved mCNN model, and perform feature evaluation on the features extracted by the model during tuning until the model iteration reaches the optimum; then use the optimal mCNN model to extract the temporal structure feature vector at multiple scales in parallel , reorganized into short-term feature sequences in chronological order, and constructed a temporal structure feature sequence set; finally, an HMM model was constructed and trained, and an offline vibration event HMM model library was constructed as a classifier for vibration source identification. The invention solves the problems that the prior art cannot simultaneously extract the local structural features and time sequence features of the distributed optical fiber vibration signal, and the vibration source identification accuracy and model generalization ability are low.

Description

technical field [0001] The invention belongs to the field of optical fiber sensing signal processing, and relates to a distributed optical fiber vibration signal feature extraction and identification method. Background technique [0002] Distributed optical fiber vibration sensing system based on phase-sensitive optical time domain reflectometry technology, using optical fiber as the sensing and transmission medium, can realize continuous monitoring of long-distance and wide-range vibration signals, with long monitoring distance, high sensitivity and environmental tolerance It is widely used in perimeter security, oil and gas pipeline safety, structural health monitoring and other civil and national defense facilities safety monitoring fields. [0003] In the practical application of long-distance monitoring, the environment is complex and changeable, multiple vibration sources are mixed, and the real-time requirements are high, which increases the difficulty of identifying ...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method for feature extraction and identification of distributed optical fiber vibration signals, which solves the problems of the prior art that cannot simultaneously extract the local structural features and timing features of distributed optical fiber vibration signals, the accuracy of vibration source identification and the model. The problem of low generalization ability

Images