A Distributed Optical Fiber Sensor Vibration Signal Classification Method and Recognition Classification System

Abstract

The invention discloses a vibration signal classification method of a distributed optical fiber sensor and an identification and classification system. The method is as follows: first, an optical fiber sensing system is used to obtain a vibration signal acting on an optical cable; the optical fiber vibration signal is preprocessed; Short-term energy and short-term zero-crossing rate; set dual thresholds for short-term energy and short-term zero-crossing rate, if the threshold is exceeded, the valid data segment is extracted and judged as a disturbance event; the time-frequency domain of the optical fiber vibration signal is plotted. Spectrogram; extract the Mel frequency cepstral coefficient of the optical fiber vibration signal; establish a deep learning recognition model based on the Mel frequency cepstral coefficient of the disturbance event signal and the time-frequency domain spectrogram; based on the spectrogram and the Mel frequency in the time-frequency domain of the vibration signal The two features of the cepstral coefficient are matched with the deep learning recognition model to determine the type of optical fiber vibration signal. The invention realizes feature extraction of optical fiber sensing signals, accurate identification and classification, and solves the technical problem of low identification and classification accuracy of optical fiber sensing intrusion signals.

Description

technical field [0001] The invention belongs to the technical field of signal detection, and in particular relates to a vibration signal classification method and identification classification system of a distributed optical fiber sensor. Background technique [0002] Distributed optical fiber sensing disturbance detection technology has increasingly become the mainstream technology of intrusion behavior security monitoring due to its strong anti-electromagnetic interference ability, high sensitivity, large dynamic range, easy to quickly capture and high-precision monitoring of dynamic change information and other technical advantages. For example, in oil pipelines, the distributed optical fiber disturbance monitoring system is used to replace manual patrols to detect oil stealing, oil leakage, and pipeline damage in time, and intelligently manage pipelines to ensure pipeline safety. In the field of security, due to the concealment and anti-electromagnetic properties of opti...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the above problems, the present invention proposes a method and system for identifying optical fiber sensing vibration signals, which realizes feature extraction of optical fiber sensing signals, accurate identification and classification, and solves the technology of low accuracy in identification and classification of optical fiber sensing intrusion signals question

Images