Distributed optical fiber vibration signal mode classification method and system based on neural network

Abstract

The invention discloses a distributed optical fiber vibration signal mode classification method based on a neural network, and the method comprises the steps: obtaining alarm real-time data transmitted by a sensor, and carrying out the low-pass filtering of the data, so as to eliminate high-frequency noise; positioning an alarm point according to the characteristics of the signal, and intercepting a signal segment which has a fixed length and meets the input length requirement of the neural network according to a positioning result; carrying out normalization processing on the signal segments and then inputting the signal segments into a neural network; and training the network by adopting a batch stochastic gradient descent algorithm to obtain a final neural network model and parameters for judgment. According to the invention, after the distributed optical fiber sensor generates the alarm signal, the reason for causing the alarm can be judged in real time, so that a manager can quickly position the place and reason of the alarm and quickly make corresponding response, and the method is simple and suitable for general popularization. The invention further discloses a distributed optical fiber vibration signal mode classification system based on the neural network.

Description

technical field [0001] The invention belongs to the technical field of signal processing, and more specifically relates to a neural network-based distributed optical fiber vibration signal mode classification method and system. Background technique [0002] Distributed optical fiber vibration monitoring system has been widely used in the fields of tunnel corridor security positioning, power system cable monitoring, information pipeline online monitoring, high-speed railway security system, long-distance pipeline security monitoring, key facility perimeter security, etc. [0003] In the existing distributed optical fiber vibration monitoring scheme, Tian Miao et al. used radial basis function combined with empirical mode decomposition to realize the recognition of four events: shearing, shaking, knocking and climbing, and the average recognition rate reached 85.75%; Li Zhichen et al. used short-time Fourier and singular value decomposition to realize feature extraction, and u...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the above defects or improvement needs of the prior art, the present invention provides a neural network-based distributed optical fiber vibration signal mode classification method and system, the purpose of which is to solve the existing distributed optical fiber vibration monitoring scheme due to the need for complex signal The processing method leads to the technical problem of complex system design, and because these schemes use the feature engineering method, the performance of the system depends on the effectiveness of the selected features, which leads to the technical problem of poor adaptability to different vibration categories

Images