1. A distributed vibration identification method based on a dual-channel phi-OTDR buried optical cable is characterized in that: the method comprises the following steps:

s1, wrapping the multi-core single-mode optical cable by using an HDPE sleeve, and laying the multi-core single-mode optical cable in a pipeline and a to-be-monitored area of a station;

s2, building a distributed optical fiber sensing system through pipeline design;

s3, carrying out matrix inversion, splicing and segmentation on the two-channel multi-group optical pulse echo signals into i groups of different matrixes, and designing a linear phase FIR digital filter group;

s4, selecting 6 vibration events and other events to acquire related data, labeling, filtering noise through a two-dimensional Kaiser window function, selecting characteristic frequency, and generating a final image data set through two-dimensional Fourier inversion;

s5, training a model by a back propagation method through a feature extraction network, a classifier and a loss function;

and S6, recognizing the newly-entered data through the signal processing and pattern recognition steps to obtain a vibration event and a vibration position, transmitting the vibration event and the vibration position to a linkage system module, and calling the camera by the system linkage module to synchronously intercept the image for comparison and warehousing.

2. The distributed vibration identification method for the buried optical cable based on the dual-channel phi-OTDR as claimed in claim 1, wherein: in the step S1, the multi-core single-mode optical cable wrapped by the HDPE casing is laid at a position 30cm above the pipeline to be monitored and about 1m away from the ground surface, soil is filled, and the optical cable in the reserved optical cable well is made shockproof by plastic foam and sponge; and laying the multi-core single-mode optical cable on the fence around the site or 10cm below the soil layer of the fence, wherein one end of the multi-core single-mode optical cable is connected to the distributed optical fiber sensing system, and the other end of the multi-core single-mode optical cable is connected with the pipeline optical cable.

3. The distributed vibration identification method for the buried optical cable based on the dual-channel phi-OTDR as claimed in claim 1, wherein: the step S2 of building the distributed optical fiber sensing system specifically includes: the pump laser emits coherent laser outwards through the G.652.D optical cable; the clock control module is connected with and controls the acousto-optic modulation driver; the acousto-optic modulation driver controls the acousto-optic modulator to shift frequency; the erbium-doped fiber amplifier improves the modulated light power; the backward Rayleigh scattered light returns to a channel 1 and a channel 2 of the photoelectric conversion equipment respectively; the photoelectric conversion device converts the optical signal into an analog signal.

4. The distributed vibration identification method for the buried optical cable based on the dual-channel phi-OTDR as claimed in claim 1, wherein: in the step S3, acquiring signal processing dual-channel m groups of optical pulse echoes, and turning, splicing and dividing the optical pulse echoes into i groups of different matrixes; according to the actual conditions of different detection areas, an m-1 order linear phase FIR digital credit filter bank which meets the symmetric conditions in the corresponding area is designed, and the corresponding matrix passes through the filter bank.

5. The distributed vibration identification method for the buried optical cable based on the dual-channel phi-OTDR as claimed in claim 1, wherein: step S4, acquiring and processing vibration data, selecting 6 times of vibration and the like to acquire relevant data according to actual requirements of gas security and marking; and (4) after the collected data pass through the step S4, performing two-dimensional Fourier transform, designing a two-dimensional Kaiser window function according to vibration characteristics to filter noise, selecting characteristic frequency, and performing two-dimensional Fourier inverse transform to generate a digital image to obtain a final data set.

Background

With the continuous development of urban construction and the continuous laying of underground gas pipelines, the gas pipelines are widely distributed in the urban underground. Based on the safety consideration of gas pipelines, especially secondary high-pressure pipelines in urban areas, once pipeline damage or gas leakage is caused by external pipeline damage behaviors such as mechanical excavation, manual damage, ground construction and the like, huge losses of lives and properties of people can be caused. In addition, the perimeter security protection of the secondary high-pressure pipeline pressurization station is also an important link of gas security protection. The traditional pipeline safety control mainly depends on manual regular patrol of patrolmen along the line, and uninterrupted safety protection for 24 hours cannot be realized; cameras are arranged at main monitoring points of the pipeline, and security personnel still need to check the cameras on duty; for the problem of long-distance gas pipeline security protection, the traditional vibration sensor needs to be buried and laid, and the problems of large construction quantity, static electricity, missing report and the like exist in power supply and monitoring point arrangement. For station perimeter security protection, the traditional methods of leakage cable monitoring, infrared correlation and the like have the problems of high false alarm rate, live working and the like, and can not effectively meet the requirements of station static prevention and accurate detection. Aiming at the operation requirements of the gas pipe network, a long-distance distributed passive vibration monitoring and identifying method is urgently needed to realize automatic alarming and protection of pipeline dangerous operation and station invasion and guarantee the operation safety of pipe network facilities.

Disclosure of Invention

Technical problem to be solved

Aiming at the defects of the prior art, the invention provides a distributed vibration identification method based on a dual-channel phi-OTDR buried optical cable, which solves the problems in the background technology.

(II) technical scheme

In order to achieve the purpose, the invention provides the following technical scheme: a distributed vibration identification method based on a dual-channel phi-OTDR buried optical cable comprises the following steps:

s1, wrapping the multi-core single-mode optical cable by using an HDPE sleeve, and laying the multi-core single-mode optical cable in a pipeline and a to-be-monitored area of a station;

s2, building a distributed optical fiber sensing system through pipeline design;

s3, carrying out matrix inversion, splicing and segmentation on the two-channel multi-group optical pulse echo signals into i groups of different matrixes, and designing a linear phase FIR digital filter group;

s4, selecting 6 vibration events and other events to acquire related data, labeling, filtering noise through a two-dimensional Kaiser window function, selecting characteristic frequency, and generating a final image data set through two-dimensional Fourier inversion;

s5, training a model by a back propagation method through a feature extraction network, a classifier and a loss function;

and S6, recognizing the newly-entered data through the signal processing and pattern recognition steps to obtain a vibration event and a vibration position, transmitting the vibration event and the vibration position to a linkage system module, and calling the camera by the system linkage module to synchronously intercept the image for comparison and warehousing.

Preferably, in the step S1, the multi-core single-mode optical cable wrapped by the HDPE sleeve is laid at a position 30cm above the pipeline to be monitored and about 1m away from the ground surface, and is covered with soil and buried, and the optical cable in the reserved optical cable well is made shockproof by plastic foam and sponge; and laying the multi-core single-mode optical cable on the fence around the site or 10cm below the soil layer of the fence, wherein one end of the multi-core single-mode optical cable is connected to the distributed optical fiber sensing system, and the other end of the multi-core single-mode optical cable is connected with the pipeline optical cable.

Preferably, the building of the distributed optical fiber sensing system in step S2 is specifically: the pump laser emits coherent laser outwards through the G.652.D optical cable; the clock control module is connected with and controls the acousto-optic modulation driver; the acousto-optic modulation driver controls the acousto-optic modulator to shift frequency; the erbium-doped fiber amplifier improves the modulated light power; the backward Rayleigh scattered light returns to a channel 1 and a channel 2 of the photoelectric conversion equipment respectively; the photoelectric conversion device converts the optical signal into an analog signal.

Preferably, in step S3, the acquired signal processing dual-channel m groups of optical pulse echoes are subjected to matrix inversion, splicing, and segmentation into i groups of different matrices; according to the actual conditions of different detection areas, an m-1 order linear phase FIR digital credit filter bank which meets the symmetric conditions in the corresponding area is designed, and the corresponding matrix passes through the filter bank.

Preferably, in the step S4, the vibration data is acquired and processed, and according to the actual demand of gas security, the vibration time and other time in step 6 are selected to acquire relevant data and labeled; and (4) after the collected data pass through the step S4, performing two-dimensional Fourier transform, designing a two-dimensional Kaiser window function according to vibration characteristics to filter noise, selecting characteristic frequency, and performing two-dimensional Fourier inverse transform to generate a digital image to obtain a final data set.

(III) advantageous effects

The invention provides a distributed vibration identification method based on a dual-channel phi-OTDR underground optical cable, which has the following beneficial effects:

aiming at the dual security requirements of a station and a pipeline in the operation of the existing gas pipe network, the invention completes the bidirectional pipe network monitoring of the station through two channels and utilizes the distributed characteristic of optical fiber sensing to realize the monitoring, positioning and identification of a single channel aiming at different security incidents of different scenes (the station and the gas pipeline) and different areas; the acquired data information can call other sensing equipment to perform corresponding operation through the linkage module, only the compatible access of the existing sensing equipment can be ensured, and the data can be reported to other security platforms through the trusted interface, so that the requirement of intelligent pipe network safety monitoring is met.

Drawings

FIG. 1 is a schematic diagram of the present invention;

fig. 2 is a schematic diagram of a vibration signal pattern recognition process according to the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

As shown in fig. 1-2, the present invention provides a technical solution: a distributed vibration identification method based on a dual-channel phi-OTDR buried optical cable comprises the following steps:

s1, laying optical fibers:

(1) laying a multi-core G.652.D single-mode optical cable protector in an HDPE sleeve at a position which is about 30cm above a secondary high-pressure pipeline and is about 1m away from the ground surface, covering soil and burying, reserving an optical cable in an optical cable well along the line, placing the optical cable at the bottom of the well to avoid suspension, and making the optical cable and an optical fiber splicing box shockproof by using plastic foam and sponge;

(2) laying a multi-core G.652.D single-mode optical cable on the surrounding fence of the station or 10cm below the soil layer of the fence, connecting one end of the multi-core G.652.D single-mode optical cable into the distributed optical fiber sensing system, and connecting the other end of the multi-core G.652.D single-mode optical cable with a pipeline optical cable

S2, building a distributed optical fiber sensing system, and using a pump laser as a seed light source to emit coherent laser outwards through a G.652.D optical cable. And a control instruction is issued to the acousto-optic modulation driver through the clock control module, and the acousto-optic modulation driver completes frequency shift control on the acousto-optic modulator. The modulated light is used for improving the light power through the erbium-doped fiber amplifier, then the modulated light is divided into two beams through the fiber coupler, and the two beams of the modulated light are respectively injected into two G.652.D optical cables through the circulator, so that vibration events of scenes in different directions and different areas of a monitoring station are correspondingly monitored. Backward Rayleigh scattered light is reflected by A₀ exp(j(2πFt₁+θ))、A₁ exp(j(2πFt₁+ theta)) and the respective circulators return to the channel 1 and the channel 2 of the photoelectric conversion device, respectively, and the optical signals are converted into analog signals through the photoelectric conversion device corresponding to the respective positions of the monitored scenest₁Representing the fast time domain, t₂Represents the slow time domain;

s3, obtaining two groups of data matrixes after the two-channel m groups of optical pulse echoes are subjected to equalization along columns by the analog-to-digital converter and the data acquisition module:

flipping the first set of matrices horizontally:

and splicing with a second group of matrixes according to rows:

according to different monitoring areas, such as stations, road crossing pipelines, soil buried pipelines and the like, the BETA is divided into different matrixes:

according to the pulse emission frequency and the signal-to-noise ratio conditions of different monitoring areas, an m-1 order linear phase FIR digital I-type band-pass filter group (h) with a corresponding area meeting the symmetry condition is designed₁[m] h₂[m] … h_i[m]). And (3) passing the corresponding matrix through the corresponding filter according to columns to obtain a filtered matrix:

s4, selecting crossing fence monitoring vibration events based on station perimeter security and manual excavation, mechanical excavation, vehicle rolling, gas leakage, directional drilling vibration and other events based on pipeline perimeter security requirements according to actual gas security requirements, acquiring relevant data, labeling corresponding labels, and repeating the step S3 to obtain an original data set;

s5, distance-time domain BETA of original dataset_i(x, t) obtaining a frequency-wavenumber domain F (F, k) through two-dimensional Fourier, namely:

determining alpha according to the frequency-wavenumber domain characteristics of different vibration events, and designing a two-dimensional Kessel window:

filtering noise, selecting characteristic frequency, generating a digital image through two-dimensional Fourier inverse transformation to obtain a final data set, and dividing the data set into a training set, a testing set and a verification set;

and S6, designing a feature extraction network, taking the position coordinates of the image and the vibration event generated in the image as input, taking a one-dimensional array (comprising the event type and the relative position of the image) as output, and taking the loss function as a weighted sum of the object identification accuracy and the relative position deviation. Training the data of the training set and the test set by using a back propagation method, and testing the trained model by using the verification set to obtain performance parameters of precision ratio, recall ratio and relative position deviation degree;

and S7, repeatedly training the network, selecting a model with relatively good performance parameters, after a new group of data is subjected to signal processing in the first five steps, judging a vibration event and a vibration position through the pattern recognition network, transmitting the vibration data to the linkage system module, and calling the camera by the system linkage module to synchronously intercept an image, comparing and storing the image.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.