CN118010150A - Method and device for monitoring optical fiber vibration event - Google Patents
Method and device for monitoring optical fiber vibration event Download PDFInfo
- Publication number
- CN118010150A CN118010150A CN202410313294.7A CN202410313294A CN118010150A CN 118010150 A CN118010150 A CN 118010150A CN 202410313294 A CN202410313294 A CN 202410313294A CN 118010150 A CN118010150 A CN 118010150A
- Authority
- CN
- China
- Prior art keywords
- vibration
- analysis model
- event type
- frequency characteristic
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000013307 optical fiber Substances 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000012544 monitoring process Methods 0.000 title claims abstract description 21
- 239000011159 matrix material Substances 0.000 claims abstract description 120
- 238000004458 analytical method Methods 0.000 claims abstract description 87
- 238000012545 processing Methods 0.000 claims abstract description 51
- 238000007405 data analysis Methods 0.000 claims abstract description 34
- 230000009467 reduction Effects 0.000 claims abstract description 19
- 239000013598 vector Substances 0.000 claims description 51
- 238000012549 training Methods 0.000 claims description 46
- 238000012360 testing method Methods 0.000 claims description 36
- 230000015654 memory Effects 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 15
- 238000010276 construction Methods 0.000 claims description 12
- 238000010586 diagram Methods 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 239000000835 fiber Substances 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims 1
- 238000004891 communication Methods 0.000 description 14
- 230000005540 biological transmission Effects 0.000 description 5
- 238000012806 monitoring device Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000000253 optical time-domain reflectometry Methods 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000010363 phase shift Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The application discloses a method and a device for monitoring an optical fiber vibration event, which are characterized in that an initial vibration signal in reverse Rayleigh scattered light generated by an optical fiber is obtained; noise reduction processing is carried out on the initial vibration signal to obtain a first vibration signal, and a time-frequency characteristic matrix of the first vibration signal is constructed; processing a time-frequency characteristic matrix by utilizing a pre-trained vibration data analysis model to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light; constructing a graph eigenvector matrix corresponding to the first vibration signal; processing the graph eigenvector matrix by using a pre-trained vibration graph analysis model to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light; and determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type. The scheme can improve the accuracy of vibration event determination, so that subsequent staff can take corresponding measures according to the determined vibration event type.
Description
Technical Field
The application relates to the technical field of optical fiber vibration monitoring, in particular to a method and a device for monitoring an optical fiber vibration event.
Background
An optical fiber line is a transmission medium for optical signals that can be transmitted from a transmitter to a receiver with as little attenuation and pulse stretching as possible. Optical fibers are often located in areas such as machine rooms, cable wells, risers, etc., and therefore there are often various events in the vicinity of the optical fibers, such as human activity, vehicle transit, etc.
However, no matter the artificial movement, the vehicle passes, or the ground crack, the pipeline crack, etc., a certain degree of vibration or sound wave is generated and then transmitted to the optical fiber, so that the optical fiber generates reverse Rayleigh scattered light, thereby affecting the normal transmission work of the optical fiber circuit. It is therefore necessary to monitor the optical fiber and to select different modes of handling for different vibration event types.
Disclosure of Invention
In view of the above, the present application provides a method and an apparatus for monitoring vibration event of optical fiber, which can accurately monitor the real vibration event type for the problem that the normal transmission work of the optical fiber line is affected by the reverse rayleigh scattering light generated by the optical fiber.
In order to achieve the above object, the following schemes are proposed:
in a first aspect, a method for monitoring a fiber vibration event includes:
acquiring an initial vibration signal in reverse Rayleigh scattered light generated by an optical fiber;
noise reduction processing is carried out on the initial vibration signal to obtain a first vibration signal, and a time-frequency characteristic matrix of the first vibration signal is constructed;
Processing the time-frequency characteristic matrix by utilizing a pre-trained vibration data analysis model to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light;
constructing a graph eigenvector matrix corresponding to the first vibration signal;
Processing the graph eigenvector matrix by using a pre-trained vibration graph analysis model to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light;
and determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type.
Preferably, the constructing the time-frequency characteristic matrix of the initial vibration signal includes:
Determining time-frequency characteristic information of the first vibration signal;
Calculating each time-frequency characteristic vector of the first vibration signal based on the time-frequency characteristic information;
Constructing each vector matrix corresponding to each time-frequency characteristic vector;
And combining the vector matrixes to obtain a time-frequency characteristic matrix.
Preferably, the constructing a graph eigenvector matrix corresponding to the first vibration signal includes:
Establishing a time-frequency characteristic image corresponding to the first vibration signal;
compensating the time-frequency characteristic image by adopting a neighborhood mean value method to obtain a compensated image;
and converting the compensation image into a graph eigenvector matrix.
Preferably, the training process of the vibration data analysis model includes:
constructing a first initial analysis model;
collecting vibration signals in reverse Rayleigh scattered light generated by various optical fibers, and summarizing the vibration signals into a vibration signal sample;
performing first pretreatment on the vibration signal sample to obtain a time-frequency characteristic matrix sample;
Splitting the time-frequency characteristic matrix sample into a first training set and a first testing set;
Inputting the first training set into the first initial analysis model, and training the first initial analysis model with the aim of minimizing a pre-established first loss function;
testing the trained first initial analysis model by using the first test set to obtain a first test result;
Judging whether the first test result meets a preset first requirement or not, if so, taking the trained first initial analysis model as a vibration data analysis model;
If not, returning to the step of executing the first training set to input the first training set into the first initial analysis model, and training the first initial analysis model with the preset first loss function minimized as a target until a first test result meets the first requirement.
Preferably, the training process of the vibration map analysis model includes:
Constructing a second initial analysis model;
collecting vibration signals in reverse Rayleigh scattered light generated by various optical fibers, and summarizing the vibration signals into a vibration signal sample;
performing second preprocessing on the vibration signal sample to obtain a graph eigenvector matrix sample;
splitting the graph feature vector matrix sample into a second training set and a second testing set;
inputting the second training set into the second initial analysis model, and training the second initial analysis model with the aim of minimizing a pre-established second loss function;
Testing the trained second initial analysis model by using the second test set to obtain a second test result;
judging whether the second test result meets a preset second requirement, if so, taking the trained second initial analysis model as a vibration diagram analysis model;
if not, returning to the step of executing the second training set to input the second training set into the second initial analysis model, and training the second initial analysis model with the aim of minimizing the second loss function established in advance until the second test result meets the second requirement.
Preferably, the vibration data analysis model comprises an input layer, a long-term and short-term memory network layer, a full-connection layer and an output layer;
The output end of the input layer is connected with the input end of the long-short-period memory network layer, the output end of the long-short-period memory network layer is connected with the input end of the full-connection layer, and the output end of the full-connection layer is connected with the input end of the output layer.
Preferably, the vibration map analysis model comprises an input layer, a map convolution neural network layer, a full connection layer and an output layer;
the output end of the input layer is connected with the input end of the graph roll-up neural network layer, the output end of the graph roll-up neural network layer is connected with the input end of the full-connection layer, and the output end of the full-connection layer is connected with the input end of the output layer.
Preferably, the determining a target vibration event type corresponding to the reverse rayleigh scattered light according to the first vibration event type and the second vibration event type includes:
Judging whether the first vibration event type is the same as the second vibration event type;
if yes, determining the first vibration event type or the second vibration event type as a target vibration event type;
If not, returning to the step of executing the noise reduction processing on the initial vibration signal to obtain a first vibration signal and constructing a time-frequency characteristic matrix of the first vibration signal until the type of the first vibration event is the same as the type of the second vibration event.
In a second aspect, a monitoring device for a fiber vibration event, comprising:
the initial vibration signal acquisition module is used for acquiring an initial vibration signal in the reverse Rayleigh scattered light generated by the optical fiber;
The time-frequency characteristic matrix construction module is used for carrying out noise reduction treatment on the initial vibration signal to obtain a first vibration signal and constructing a time-frequency characteristic matrix of the first vibration signal;
the first vibration event type determining module is used for processing the time-frequency characteristic matrix by utilizing a pre-trained vibration data analysis model so as to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light;
The diagram feature vector matrix construction module is used for constructing a diagram feature vector matrix corresponding to the first vibration signal;
the second vibration event type determining module is used for processing the graph eigenvector matrix by utilizing a pre-trained vibration graph analysis model so as to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light;
And the target vibration event type determining module is used for determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type.
Preferably, the time-frequency characteristic matrix construction module includes:
The time-frequency characteristic information determining module is used for determining the time-frequency characteristic information of the first vibration signal;
The time-frequency characteristic vector calculation module is used for calculating each time-frequency characteristic vector of the first vibration signal based on the time-frequency characteristic information;
the vector matrix construction module is used for constructing each vector matrix corresponding to each time-frequency characteristic vector;
And the combination module is used for combining the vector matrixes to obtain a time-frequency characteristic matrix.
According to the technical scheme, the method and the device can obtain the initial vibration signal in the reverse Rayleigh scattered light generated by the optical fiber; noise reduction processing is carried out on the initial vibration signal to obtain a first vibration signal, and a time-frequency characteristic matrix of the first vibration signal is constructed; processing the time-frequency characteristic matrix by utilizing a pre-trained vibration data analysis model to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light; constructing a graph eigenvector matrix corresponding to the first vibration signal; processing the graph eigenvector matrix by using a pre-trained vibration graph analysis model to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light; and determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type. According to the scheme, an initial vibration signal is extracted from reverse Rayleigh scattered light, noise in the initial vibration signal is removed through noise reduction processing, a time-frequency characteristic matrix and a graph characteristic vector matrix of a first vibration signal are processed through a pre-trained vibration data analysis model and a pre-trained vibration graph analysis model respectively, a first vibration event type and a second vibration event type corresponding to the reverse Rayleigh scattered light are determined respectively, a target vibration event type is finally determined from the vibration event types determined by the two models, the event type is judged through two different matrixes, the misjudgment probability can be reduced, and the accuracy of vibration event determination can be improved through the analysis model, so that follow-up workers can take corresponding measures according to the determined vibration event types.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is an alternative flow chart of a method for monitoring an optical fiber vibration event according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a device for monitoring an optical fiber vibration event according to an embodiment of the present application;
Fig. 3 is a schematic structural diagram of a monitoring device for an optical fiber vibration event according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
An optical fiber line is a transmission medium for optical signals that can be transmitted from a transmitter to a receiver with as little attenuation and pulse stretching as possible. Optical fibers are often located in areas such as machine rooms, cable wells, risers, etc., and therefore there are often various events in the vicinity of the optical fibers, such as human activity, vehicle transit, etc.
However, no matter the artificial movement, the vehicle passes, or the ground crack, the pipeline crack, etc., a certain degree of vibration or sound wave is generated and then transmitted to the optical fiber, so that the optical fiber generates reverse Rayleigh scattered light, thereby affecting the normal transmission work of the optical fiber circuit. It is therefore necessary to monitor the optical fiber and to select different modes of handling for different vibration event types.
The existing base station has certain limitations when processing complex scene communication tasks, such as difficulty in meeting the requirement of rapid signal processing, the use of low-cost hardware (such as a low-energy-consumption low-resolution analog-to-digital converter) can introduce additional nonlinear defects, a highly robust receiving processing algorithm is required to be used, meanwhile, as the dimension of communication signals to be processed is increased increasingly, the calculation complexity and time delay of signal processing are greatly increased, and the problems adopt the traditional communication algorithm to easily cause calculation to reach bottleneck, so that the calculation efficiency is reduced; on the other hand, conventional communication systems, which are embodied in limited modular structure communication systems and are constructed in a divide-and-conquer manner, consist of a series of artificially defined signal processing modules, such as coding, modulation and detection. Researchers have typically achieved better system performance by optimizing each module independently, but researchers have been trying to optimize the algorithm of each processing module for many years and have been successful in practice, but have not guaranteed that the overall communication system achieves optimal performance.
Based on the problems of the base station, the capability of processing the vibration information of the optical fiber is relatively weak, for example, in a base station system based on the phi-OTDR system, the phi-OTDR system obtains the characteristics and the positions of events happening in the optical fiber by calculating the phase shift condition of the reverse Rayleigh scattered light returned by the scattering points of the incident light at different positions in the optical fiber, so that various vibration conditions generated on an optical fiber line can be returned to the base station for processing analysis in a form of the reverse Rayleigh scattered light, the data processing capacity is large, and the conventional base station cannot bear or process a large amount of data, so that the capability of processing the data is greatly reduced, and the judgment precision and the judgment efficiency of the vibration events are seriously affected.
The embodiment of the invention provides a method for monitoring an optical fiber vibration event, which can be applied to various computer terminals or intelligent terminals, wherein an execution subject of the method can be a processor or a server of the computer terminal or the intelligent terminal, and a flow chart of the method is shown in fig. 1, and specifically comprises the following steps:
s1: an initial vibration signal in the reverse Rayleigh scattered light generated by the optical fiber is acquired.
Various vibration signal collectors can be utilized to collect vibration signals in the reverse Rayleigh scattered light generated by the optical fiber as initial vibration signals.
The application is suitable for vibration signals generated at one position of the optical fiber.
S2: and carrying out noise reduction treatment on the initial vibration signal to obtain a first vibration signal, and constructing a time-frequency characteristic matrix of the first vibration signal.
Because the optical fibers are located at different positions and in different environments, noise exists in the obtained initial vibration signal, if the vibration event type is directly determined by processing the initial vibration signal, the accuracy of determination is greatly reduced, and therefore, the noise reduction or denoising processing is firstly performed on the initial vibration signal, for example, a filter, a statistical method or a wavelet denoising method is adopted to filter the noise signal in the initial vibration signal.
And after denoising is completed, obtaining a first vibration signal, and constructing a time-frequency characteristic matrix of the first vibration signal.
S3: and processing the time-frequency characteristic matrix by using a pre-trained vibration data analysis model to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light.
According to the application, a vibration data analysis model is trained in advance, an artificial intelligent neural network model can be used as the model, and the time-frequency characteristic matrix is processed by the model, so that the analysis efficiency and quality can be improved, and the first vibration event type can be obtained quickly, timely and accurately.
S4: and constructing a graph eigenvector matrix corresponding to the first vibration signal.
S5: and processing the graph eigenvector matrix by using a pre-trained vibration graph analysis model to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light.
Besides constructing a time-frequency characteristic matrix of the first vibration signal and analyzing the time-frequency characteristic matrix, in order to ensure higher analysis accuracy, the application also pre-trains a vibration graph analysis model, constructs a graph characteristic vector matrix corresponding to the first vibration signal, and respectively determines the vibration event type corresponding to the reverse Rayleigh scattered light from two angles, namely the time-frequency characteristic matrix and the graph characteristic vector matrix. Wherein the graph eigenvector matrix is processed using a pre-trained vibration graph analysis model to obtain a second vibration event type.
S6: and determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type.
And finally judging the vibration event types analyzed by the two models, and determining the target vibration event type corresponding to the reverse Rayleigh scattered light, so that the condition that the whole analysis result is influenced due to the analysis error of any one model can be avoided.
Common vibration event types include: the personnel walk, run, the vehicles pass, move, ground, crack, the pipeline is installed, disassembled, broken and the like.
According to the technical scheme, the method and the device can obtain the initial vibration signal in the reverse Rayleigh scattered light generated by the optical fiber; noise reduction processing is carried out on the initial vibration signal to obtain a first vibration signal, and a time-frequency characteristic matrix of the first vibration signal is constructed; processing the time-frequency characteristic matrix by utilizing a pre-trained vibration data analysis model to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light; constructing a graph eigenvector matrix corresponding to the first vibration signal; processing the graph eigenvector matrix by using a pre-trained vibration graph analysis model to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light; and determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type. According to the scheme, an initial vibration signal is extracted from reverse Rayleigh scattered light, noise in the initial vibration signal is removed through noise reduction processing, a time-frequency characteristic matrix and a graph characteristic vector matrix of a first vibration signal are processed through a pre-trained vibration data analysis model and a pre-trained vibration graph analysis model respectively, a first vibration event type and a second vibration event type corresponding to the reverse Rayleigh scattered light are determined respectively, a target vibration event type is finally determined from the vibration event types determined by the two models, the event type is judged through two different matrixes, the misjudgment probability can be reduced, and the accuracy of vibration event determination can be improved through the analysis model, so that follow-up workers can take corresponding measures according to the determined vibration event types.
In the method provided by the embodiment of the invention, a process of constructing the time-frequency characteristic matrix of the initial vibration signal is specifically described as follows:
Determining time-frequency characteristic information of the first vibration signal;
Calculating each time-frequency characteristic vector of the first vibration signal based on the time-frequency characteristic information;
Constructing each vector matrix corresponding to each time-frequency characteristic vector;
And combining the vector matrixes to obtain a time-frequency characteristic matrix.
Specifically, in order to more intuitively and accurately analyze the vibration event type, the application firstly analyzes the characteristics of the first vibration signal to determine the time-frequency characteristic information of the first vibration signal, wherein the time-frequency characteristic information is the statistical characteristic extracted from the time domain and the frequency domain of the first vibration signal, and the information can describe the basic properties of the first vibration signal, such as probability distribution, power spectrum, autocorrelation coefficient and the like.
After the time-frequency characteristic information is obtained, one or more time-frequency characteristic vectors of the first vibration signal are calculated according to the information, the time-frequency characteristic information is displayed in a vector form, then a vector matrix corresponding to each time-frequency characteristic vector is constructed, and the time-frequency characteristic vectors are combined and summarized to form a time-frequency characteristic matrix. Compared with the method for directly processing the first vibration signal, the method has the advantages that the efficiency is higher when the time-frequency characteristic matrix is processed by utilizing the pre-constructed vibration data analysis model, the time-frequency characteristic matrix directly indicates the characteristic information of the first vibration signal, and the model can obtain more accurate vibration event types according to the time-frequency characteristic matrix.
In the application, a graph eigenvector matrix corresponding to the first vibration signal is constructed, which specifically comprises the following steps:
Establishing a time-frequency characteristic image corresponding to the first vibration signal;
compensating the time-frequency characteristic image by adopting a neighborhood mean value method to obtain a compensated image;
and converting the compensation image into a graph eigenvector matrix.
Specifically, the time-frequency characteristic image can specifically and intuitively show the frequency change information of the first vibration signal, then the time-frequency characteristic image is compensated by adopting a neighborhood mean method, the compensated image is converted into a graph characteristic vector matrix after compensation, and then the graph characteristic vector matrix is input into a vibration graph analysis model, so that the vibration event type obtained by analyzing by taking the graph characteristic vector matrix as a starting point can be obtained.
The training process of the vibration data analysis model and the training process of the vibration map analysis model in the present application will be specifically described below, respectively.
And (one) a vibration data analysis model.
Constructing a first initial analysis model;
collecting vibration signals in reverse Rayleigh scattered light generated by various optical fibers, and summarizing the vibration signals into a vibration signal sample;
performing first pretreatment on the vibration signal sample to obtain a time-frequency characteristic matrix sample;
Splitting the time-frequency characteristic matrix sample into a first training set and a first testing set;
Inputting the first training set into the first initial analysis model, and training the first initial analysis model with the aim of minimizing a pre-established first loss function;
testing the trained first initial analysis model by using the first test set to obtain a first test result;
Judging whether the first test result meets a preset first requirement or not, if so, taking the trained first initial analysis model as a vibration data analysis model;
If not, returning to the step of executing the first training set to input the first training set into the first initial analysis model, and training the first initial analysis model with the preset first loss function minimized as a target until a first test result meets the first requirement.
And (II) a vibration map analysis model.
Constructing a second initial analysis model;
collecting vibration signals in reverse Rayleigh scattered light generated by various optical fibers, and summarizing the vibration signals into a vibration signal sample;
performing second preprocessing on the vibration signal sample to obtain a graph eigenvector matrix sample;
splitting the graph feature vector matrix sample into a second training set and a second testing set;
inputting the second training set into the second initial analysis model, and training the second initial analysis model with the aim of minimizing a pre-established second loss function;
Testing the trained second initial analysis model by using the second test set to obtain a second test result;
judging whether the second test result meets a preset second requirement, if so, taking the trained second initial analysis model as a vibration diagram analysis model;
if not, returning to the step of executing the second training set to input the second training set into the second initial analysis model, and training the second initial analysis model with the aim of minimizing the second loss function established in advance until the second test result meets the second requirement.
Specifically, in order to enable the vibration data analysis model and the vibration map analysis model to cope with various situations that may be faced by different areas, different optical fibers and optical fibers, when a sample is collected, the vibration data analysis model and the vibration map analysis model may be used for carrying out area division according to situations such as on-site arrangement of an optical fiber line and the effective monitoring range of communication terminals distributed and arranged on the optical fiber line, so as to collect vibration signals in reverse Rayleigh scattered light generated by the optical fibers in each area respectively. And vibration signals respectively generated by natural factors and artificial factors can be distinguished to establish different vibration signal samples. The vibration data analysis model and the vibration map analysis model are trained by utilizing different vibration signal samples, parameters, structures and the like of the model can be flexibly adjusted, so that the model is trained to be optimal, the trained model can cope with various complex conditions, and the model is prevented from being over-fitted under one scene, so that the model precision is improved.
Optionally, vibration signals caused by different areas or different factors can be separately collected, so that a model can be trained in a targeted manner, for example, if some areas are remote areas or areas with fewer vibration occurrence areas, the sample collection amount of the areas can be reduced, the scale of the model can be simplified, the training process of the model is shortened, the training efficiency is improved, meanwhile, the setting cost of hardware facilities can be reduced, and the trained model can be processed and analyzed in a targeted manner when meeting real input data of a specific area, so that the accuracy of analysis is greatly increased.
In the actual monitoring of the optical fiber vibration event, if the vibration data analysis model and the vibration map analysis model in the base station covered by the corresponding area cannot identify the time-frequency characteristic matrix or the map characteristic vector matrix corresponding to the first vibration signal in the reverse Rayleigh scattered light generated by the optical fiber, the base station sent to the adjacent area can be remodulated, and the vibration data analysis model and the vibration map analysis model corresponding to the base station in the adjacent area are used for analysis.
The above-described embodiments explain the training process of the vibration data analysis model and the training process of the vibration map analysis model in the present application, and the following describes the model structures of the vibration data analysis model and the vibration map analysis model in the present application.
The vibration data analysis model comprises an input layer, a Long Short-Term Memory network Layer (LSTM), a full connection layer and an output layer;
The output end of the input layer is connected with the input end of the long-short-period memory network layer, the output end of the long-short-period memory network layer is connected with the input end of the full-connection layer, and the output end of the full-connection layer is connected with the input end of the output layer.
The vibration map analysis model comprises an input layer, a map convolution neural network layer, a full connection layer and an output layer;
the output end of the input layer is connected with the input end of the graph roll-up neural network layer, the output end of the graph roll-up neural network layer is connected with the input end of the full-connection layer, and the output end of the full-connection layer is connected with the input end of the output layer.
The following embodiments describe in detail a process of determining a target vibration event type corresponding to the reverse rayleigh scattered light according to the first vibration event type and the second vibration event type in the present application.
Judging whether the first vibration event type is the same as the second vibration event type;
if yes, determining the first vibration event type or the second vibration event type as a target vibration event type;
If not, returning to the step of executing the noise reduction processing on the initial vibration signal to obtain a first vibration signal and constructing a time-frequency characteristic matrix of the first vibration signal until the type of the first vibration event is the same as the type of the second vibration event.
In the scheme, the method and the device for determining the vibration event type through the two models can ensure the certainty and the accuracy of the vibration event type, and prevent one model from influencing the final recognition result due to the recognition error.
Corresponding to the method described in fig. 1, the embodiment of the present invention further provides a device for monitoring an optical fiber vibration event, which is used for implementing the method in fig. 1, where the device for monitoring an optical fiber vibration event provided in the embodiment of the present invention may be introduced in a computer terminal or various mobile devices with reference to fig. 2, and as shown in fig. 2, the device may include:
an initial vibration signal acquisition module 10, configured to acquire an initial vibration signal in the reverse rayleigh scattered light generated by the optical fiber;
the time-frequency characteristic matrix construction module 20 is used for carrying out noise reduction treatment on the initial vibration signal to obtain a first vibration signal, and constructing a time-frequency characteristic matrix of the first vibration signal;
a first vibration event type determining module 30, configured to process the time-frequency feature matrix by using a pre-trained vibration data analysis model, so as to obtain a first vibration event type corresponding to the reverse rayleigh scattered light;
A graph eigenvector matrix construction module 40, configured to construct a graph eigenvector matrix corresponding to the first vibration signal;
a second vibration event type determining module 50 for processing the graph eigenvector matrix using a pre-trained vibration graph analysis model to obtain a second vibration event type corresponding to the reverse rayleigh scattered light;
A target vibration event type determining module 60, configured to determine a target vibration event type corresponding to the reverse rayleigh scattered light according to the first vibration event type and the second vibration event type.
According to the technical scheme, the method and the device can obtain the initial vibration signal in the reverse Rayleigh scattered light generated by the optical fiber; noise reduction processing is carried out on the initial vibration signal to obtain a first vibration signal, and a time-frequency characteristic matrix of the first vibration signal is constructed; processing the time-frequency characteristic matrix by utilizing a pre-trained vibration data analysis model to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light; constructing a graph eigenvector matrix corresponding to the first vibration signal; processing the graph eigenvector matrix by using a pre-trained vibration graph analysis model to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light; and determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type. According to the scheme, an initial vibration signal is extracted from reverse Rayleigh scattered light, noise in the initial vibration signal is removed through noise reduction processing, a time-frequency characteristic matrix and a graph characteristic vector matrix of a first vibration signal are processed through a pre-trained vibration data analysis model and a pre-trained vibration graph analysis model respectively, a first vibration event type and a second vibration event type corresponding to the reverse Rayleigh scattered light are determined respectively, a target vibration event type is finally determined from the vibration event types determined by the two models, the event type is judged through two different matrixes, the misjudgment probability can be reduced, and the accuracy of vibration event determination can be improved through the analysis model, so that follow-up workers can take corresponding measures according to the determined vibration event types.
In one example, the time-frequency feature matrix construction module 20 may include:
The time-frequency characteristic information determining module is used for determining the time-frequency characteristic information of the first vibration signal;
The time-frequency characteristic vector calculation module is used for calculating each time-frequency characteristic vector of the first vibration signal based on the time-frequency characteristic information;
the vector matrix construction module is used for constructing each vector matrix corresponding to each time-frequency characteristic vector;
And the combination module is used for combining the vector matrixes to obtain a time-frequency characteristic matrix.
Further, the embodiment of the application provides monitoring equipment for an optical fiber vibration event. Optionally, fig. 3 shows a block diagram of a hardware structure of a monitoring device for a fiber vibration event, and referring to fig. 3, the hardware structure of the monitoring device for a fiber vibration event may include: at least one processor 01, at least one communication interface 02, at least one memory 03 and at least one communication bus 04.
In the embodiment of the present application, the number of the processor 01, the communication interface 02, the memory 03 and the communication bus 04 is at least one, and the processor 01, the communication interface 02 and the memory 03 complete communication with each other through the communication bus 04.
The processor 01 may be a central processing unit CPU, or an Application-specific integrated Circuit ASIC (Application SPECIFIC INTEGRATED Circuit), or one or more integrated circuits configured to implement embodiments of the present invention, or the like.
The memory 03 may include a high-speed RAM memory, and may further include a nonvolatile memory (non-volatile memory) or the like, such as at least one magnetic disk memory.
The memory stores a program, and the processor can call the program stored in the memory, and the program is used for executing the following monitoring method of the optical fiber vibration event, and the method comprises the following steps:
acquiring an initial vibration signal in reverse Rayleigh scattered light generated by an optical fiber;
noise reduction processing is carried out on the initial vibration signal to obtain a first vibration signal, and a time-frequency characteristic matrix of the first vibration signal is constructed;
Processing the time-frequency characteristic matrix by utilizing a pre-trained vibration data analysis model to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light;
constructing a graph eigenvector matrix corresponding to the first vibration signal;
Processing the graph eigenvector matrix by using a pre-trained vibration graph analysis model to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light;
and determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type.
Alternatively, the refinement and expansion functions of the program may be described with reference to a method of monitoring a fiber vibration event in a method embodiment.
The embodiment of the application also provides a storage medium, which can store a program suitable for being executed by a processor, and when the program runs, the device where the storage medium is controlled to execute the following monitoring method of the optical fiber vibration event, comprising the following steps:
acquiring an initial vibration signal in reverse Rayleigh scattered light generated by an optical fiber;
noise reduction processing is carried out on the initial vibration signal to obtain a first vibration signal, and a time-frequency characteristic matrix of the first vibration signal is constructed;
Processing the time-frequency characteristic matrix by utilizing a pre-trained vibration data analysis model to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light;
constructing a graph eigenvector matrix corresponding to the first vibration signal;
Processing the graph eigenvector matrix by using a pre-trained vibration graph analysis model to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light;
and determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type.
In particular, the storage medium may be a computer-readable storage medium, which may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM.
Alternatively, the refinement and expansion functions of the program may be described with reference to a method of monitoring a fiber vibration event in a method embodiment.
In addition, functional modules in various embodiments of the present disclosure may be integrated together to form a single portion, or each module may exist alone, or two or more modules may be integrated to form a single portion. The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored on a computer readable storage medium. Based on such understanding, the technical solution of the present disclosure may be embodied in essence or a part contributing to the prior art or a part of the technical solution, or in the form of a software product stored in a storage medium, including several instructions to cause a computer device (which may be a personal computer, a live device, or a network device, etc.) to perform all or part of the steps of the methods of the embodiments of the present disclosure.
Finally, it is further noted that relational terms such as first and second, and the like are 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. Moreover, 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. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method of monitoring a fiber vibration event, comprising:
acquiring an initial vibration signal in reverse Rayleigh scattered light generated by an optical fiber;
noise reduction processing is carried out on the initial vibration signal to obtain a first vibration signal, and a time-frequency characteristic matrix of the first vibration signal is constructed;
Processing the time-frequency characteristic matrix by utilizing a pre-trained vibration data analysis model to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light;
constructing a graph eigenvector matrix corresponding to the first vibration signal;
Processing the graph eigenvector matrix by using a pre-trained vibration graph analysis model to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light;
and determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type.
2. The method of claim 1, wherein said constructing a time-frequency characteristics matrix of said initial vibration signal comprises:
Determining time-frequency characteristic information of the first vibration signal;
Calculating each time-frequency characteristic vector of the first vibration signal based on the time-frequency characteristic information;
Constructing each vector matrix corresponding to each time-frequency characteristic vector;
And combining the vector matrixes to obtain a time-frequency characteristic matrix.
3. The method of claim 1, wherein constructing a graph eigenvector matrix corresponding to the first vibration signal comprises:
Establishing a time-frequency characteristic image corresponding to the first vibration signal;
compensating the time-frequency characteristic image by adopting a neighborhood mean value method to obtain a compensated image;
and converting the compensation image into a graph eigenvector matrix.
4. The method of claim 1, wherein the training process of the vibration data analysis model comprises:
constructing a first initial analysis model;
collecting vibration signals in reverse Rayleigh scattered light generated by various optical fibers, and summarizing the vibration signals into a vibration signal sample;
performing first pretreatment on the vibration signal sample to obtain a time-frequency characteristic matrix sample;
Splitting the time-frequency characteristic matrix sample into a first training set and a first testing set;
Inputting the first training set into the first initial analysis model, and training the first initial analysis model with the aim of minimizing a pre-established first loss function;
testing the trained first initial analysis model by using the first test set to obtain a first test result;
Judging whether the first test result meets a preset first requirement or not, if so, taking the trained first initial analysis model as a vibration data analysis model;
If not, returning to the step of executing the first training set to input the first training set into the first initial analysis model, and training the first initial analysis model with the preset first loss function minimized as a target until a first test result meets the first requirement.
5. The method of claim 1, wherein the training process of the vibration map analysis model comprises:
Constructing a second initial analysis model;
collecting vibration signals in reverse Rayleigh scattered light generated by various optical fibers, and summarizing the vibration signals into a vibration signal sample;
performing second preprocessing on the vibration signal sample to obtain a graph eigenvector matrix sample;
splitting the graph feature vector matrix sample into a second training set and a second testing set;
inputting the second training set into the second initial analysis model, and training the second initial analysis model with the aim of minimizing a pre-established second loss function;
Testing the trained second initial analysis model by using the second test set to obtain a second test result;
judging whether the second test result meets a preset second requirement, if so, taking the trained second initial analysis model as a vibration diagram analysis model;
if not, returning to the step of executing the second training set to input the second training set into the second initial analysis model, and training the second initial analysis model with the aim of minimizing the second loss function established in advance until the second test result meets the second requirement.
6. The method of claim 1, wherein the vibration data analysis model comprises an input layer, a long-short-term memory network layer, a full connection layer, and an output layer;
The output end of the input layer is connected with the input end of the long-short-period memory network layer, the output end of the long-short-period memory network layer is connected with the input end of the full-connection layer, and the output end of the full-connection layer is connected with the input end of the output layer.
7. The method of claim 1, wherein the vibration map analysis model comprises an input layer, a map convolutional neural network layer, a fully connected layer, and an output layer;
the output end of the input layer is connected with the input end of the graph roll-up neural network layer, the output end of the graph roll-up neural network layer is connected with the input end of the full-connection layer, and the output end of the full-connection layer is connected with the input end of the output layer.
8. The method of claim 1, wherein the determining a target vibration event type corresponding to the reverse rayleigh scattered light from the first vibration event type and the second vibration event type comprises:
Judging whether the first vibration event type is the same as the second vibration event type;
if yes, determining the first vibration event type or the second vibration event type as a target vibration event type;
If not, returning to the step of executing the noise reduction processing on the initial vibration signal to obtain a first vibration signal and constructing a time-frequency characteristic matrix of the first vibration signal until the type of the first vibration event is the same as the type of the second vibration event.
9. A device for monitoring a fiber vibration event, comprising:
the initial vibration signal acquisition module is used for acquiring an initial vibration signal in the reverse Rayleigh scattered light generated by the optical fiber;
The time-frequency characteristic matrix construction module is used for carrying out noise reduction treatment on the initial vibration signal to obtain a first vibration signal and constructing a time-frequency characteristic matrix of the first vibration signal;
the first vibration event type determining module is used for processing the time-frequency characteristic matrix by utilizing a pre-trained vibration data analysis model so as to obtain a first vibration event type corresponding to the reverse Rayleigh scattered light;
The diagram feature vector matrix construction module is used for constructing a diagram feature vector matrix corresponding to the first vibration signal;
the second vibration event type determining module is used for processing the graph eigenvector matrix by utilizing a pre-trained vibration graph analysis model so as to obtain a second vibration event type corresponding to the reverse Rayleigh scattered light;
And the target vibration event type determining module is used for determining a target vibration event type corresponding to the reverse Rayleigh scattered light according to the first vibration event type and the second vibration event type.
10. The apparatus of claim 9, wherein the time-frequency feature matrix construction module comprises:
The time-frequency characteristic information determining module is used for determining the time-frequency characteristic information of the first vibration signal;
The time-frequency characteristic vector calculation module is used for calculating each time-frequency characteristic vector of the first vibration signal based on the time-frequency characteristic information;
the vector matrix construction module is used for constructing each vector matrix corresponding to each time-frequency characteristic vector;
And the combination module is used for combining the vector matrixes to obtain a time-frequency characteristic matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410313294.7A CN118010150A (en) | 2024-03-19 | 2024-03-19 | Method and device for monitoring optical fiber vibration event |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410313294.7A CN118010150A (en) | 2024-03-19 | 2024-03-19 | Method and device for monitoring optical fiber vibration event |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118010150A true CN118010150A (en) | 2024-05-10 |
Family
ID=90952323
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410313294.7A Pending CN118010150A (en) | 2024-03-19 | 2024-03-19 | Method and device for monitoring optical fiber vibration event |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118010150A (en) |
-
2024
- 2024-03-19 CN CN202410313294.7A patent/CN118010150A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111442827B (en) | Optical fiber passive online monitoring system for transformer winding vibration | |
| CN101334434B (en) | Electromagnetic environment test system for extracting electromagnetic leakage signal by utilizing wavelet transformation | |
| CN112985574B (en) | High-precision classification identification method for optical fiber distributed acoustic sensing signals based on model fusion | |
| CN111222743A (en) | Method for judging vertical offset distance and threat level of optical fiber sensing event | |
| CN113340353B (en) | Monitoring method, equipment and medium for power transmission line | |
| US11668857B2 (en) | Device, method and computer program product for validating data provided by a rain sensor | |
| CN110533115B (en) | Quantitative evaluation method for transmission characteristics of track circuit based on variational modal decomposition | |
| CN117191147A (en) | Flood discharge dam water level monitoring and early warning method and system | |
| CN111444233A (en) | Method for discovering environmental monitoring abnormal data based on duplicator neural network model | |
| CN114722480A (en) | Safety monitoring system of house building structure and building and monitoring method thereof | |
| CN114330120A (en) | 24-hour PM prediction based on deep neural network2.5Method of concentration | |
| CN111951505B (en) | Fence vibration intrusion positioning and mode identification method based on distributed optical fiber system | |
| CN117688305A (en) | Anomaly detection method and system based on improved noise reduction self-encoder | |
| CN115459868B (en) | Millimeter wave communication performance evaluation method and system in complex environment | |
| KR20220132824A (en) | Distribution facility condition monitoring system and method | |
| CN117473263A (en) | Automatic recognition method and system for frequency and damping ratio of bridge vibration monitoring | |
| CN118010150A (en) | Method and device for monitoring optical fiber vibration event | |
| CN112128950A (en) | Machine room temperature and humidity prediction method and system based on multiple model comparisons | |
| CN116823067A (en) | Method and device for determining water quality cleaning state of pipe network and electronic equipment | |
| CN116383723A (en) | Debris flow data anomaly identification method, computer equipment and medium | |
| CN116128690A (en) | Carbon emission cost value calculation method, device, equipment and medium | |
| CN115980690A (en) | Clear sky echo recognition model construction method, clear sky echo filtering method and clear sky echo filtering device | |
| CN113255137B (en) | Target object strain data processing method and device and storage medium | |
| CN112114215A (en) | Transformer aging evaluation method and system based on error back propagation algorithm | |
| CN115423221B (en) | Facility operation trend prediction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |