CN114518182B - Method and system for simultaneously extracting temperature and strain information in brillouin scattering spectrum image - Google Patents
Method and system for simultaneously extracting temperature and strain information in brillouin scattering spectrum image Download PDFInfo
- Publication number
- CN114518182B CN114518182B CN202210203784.2A CN202210203784A CN114518182B CN 114518182 B CN114518182 B CN 114518182B CN 202210203784 A CN202210203784 A CN 202210203784A CN 114518182 B CN114518182 B CN 114518182B
- Authority
- CN
- China
- Prior art keywords
- scattering spectrum
- brillouin scattering
- brillouin
- neural network
- convolutional neural
- 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.)
- Active
Links
- 238000001228 spectrum Methods 0.000 title claims abstract description 106
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 48
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 46
- 239000013307 optical fiber Substances 0.000 claims abstract description 35
- 239000011159 matrix material Substances 0.000 claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000012360 testing method Methods 0.000 claims abstract description 11
- 230000002902 bimodal effect Effects 0.000 claims description 11
- 230000003595 spectral effect Effects 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 abstract description 19
- 238000000605 extraction Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 3
- 238000001069 Raman spectroscopy Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 230000010287 polarization Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K11/00—Measuring temperature based upon physical or chemical changes not covered by groups G01K3/00, G01K5/00, G01K7/00 or G01K9/00
- G01K11/32—Measuring temperature based upon physical or chemical changes not covered by groups G01K3/00, G01K5/00, G01K7/00 or G01K9/00 using changes in transmittance, scattering or luminescence in optical fibres
- G01K11/322—Measuring temperature based upon physical or chemical changes not covered by groups G01K3/00, G01K5/00, G01K7/00 or G01K9/00 using changes in transmittance, scattering or luminescence in optical fibres using Brillouin scattering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/16—Measuring arrangements characterised by the use of optical techniques for measuring the deformation in a solid, e.g. optical strain gauge
- G01B11/18—Measuring arrangements characterised by the use of optical techniques for measuring the deformation in a solid, e.g. optical strain gauge using photoelastic elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention discloses a method and a system for simultaneously extracting temperature and strain information in a brillouin scattering spectrum image, and relates to the field of distributed optical fiber sensing. The method comprises the following steps: acquiring a brillouin frequency shift temperature coefficient and a strain coefficient corresponding to a double-peak brillouin scattering spectrum of the optical fiber with a large effective area; randomly combining a plurality of groups of double-peak Brillouin scattering spectrums to form a Brillouin scattering spectrum image matrix, wherein the Brillouin scattering spectrum image matrix is used as a training set of a convolutional neural network; training the convolutional neural network by adopting a training set, and iteratively optimizing parameters of the convolutional neural network by utilizing an error back propagation algorithm to obtain a trained convolutional neural network; and (3) taking actually measured brillouin scattering spectrum data as a test set, inputting the test set into a convolutional neural network after training, and simultaneously obtaining temperature and strain information at the output end of the network. The invention realizes simultaneous measurement of temperature and strain while ensuring measurement precision, obviously shortens data processing time and meets the requirement of rapid measurement.
Description
Technical Field
The invention relates to the technical field of distributed optical fiber sensing, in particular to a method and a system for simultaneously extracting temperature and strain information in a brillouin scattering spectrum image.
Background
The Brillouin distributed optical fiber sensing system is widely applied to health state monitoring and fault diagnosis of large building structures and equipment in the fields of electric power, petroleum, transportation and the like. The system generally obtains sensing information by measuring the brillouin frequency shift, the center frequency corresponding to the maximum amplitude of the brillouin scattering spectrum can be found by using a curve fitting method to determine the brillouin frequency shift, and then the temperature and strain information distributed along the sensing optical fiber are obtained by conversion according to the linear relation between the brillouin frequency shift and the temperature and strain. However, since the brillouin frequency shift is sensitive to temperature and strain at the same time, when the temperature and strain of a sensing optical fiber at a certain position along the line change at the same time, the brillouin frequency shift alone cannot distinguish the temperature and the strain at the same time, that is, the brillouin optical fiber sensing system has the problem of cross sensitivity of temperature and strain measurement, which limits the application of the system in special occasions.
At present, two main methods for solving the cross-sensitivity problem of temperature and strain measurement are as follows: the mixed sensing system combining the Brillouin scattering and the Raman scattering adopts special optical fibers (such as large effective area optical fibers, polarization maintaining optical fibers, multi-core optical fibers and the like) for sensing measurement. The hybrid sensing system combining the Brillouin scattering and the Raman scattering has a complex structure and high cost, so that the application of the hybrid sensing system is limited. The method for adopting the special optical fiber is characterized in that the temperature and the strain are measured simultaneously by utilizing a mode of solving an equation set according to the measured Brillouin scattering spectra with a plurality of peaks or the Brillouin frequency shift corresponding to different optical modes with different temperature and strain coefficients. However, when a special optical fiber is used to perform the measurement, the following problems still remain: taking a large effective area optical fiber as an example, for the same sensing optical fiber, the temperature and strain coefficients of corresponding Brillouin frequency shifts at different scattering spectrum peaks are very close, and larger errors are easily generated in the process of solving an equation set, so that the measurement accuracy of the temperature and the strain is reduced. In addition, the currently adopted brillouin frequency shift determining method is still mainly based on a curve fitting method, which requires long data processing time and can influence the system measurement speed.
Therefore, how to efficiently extract temperature and strain information in brillouin scattering spectrum images while ensuring measurement accuracy is a problem to be solved for those skilled in the art.
Disclosure of Invention
In view of the above, the present invention provides a method and a system for simultaneously extracting temperature and strain information in a brillouin scattering spectrum image, so as to solve the problems set forth in the background art.
In order to achieve the above purpose, the present invention adopts the following technical scheme: in one aspect, a method for simultaneously extracting temperature and strain information in a brillouin scattering spectrum image is provided, which specifically includes the following steps:
acquiring a brillouin frequency shift temperature coefficient and a strain coefficient corresponding to a double-peak brillouin scattering spectrum of the optical fiber with a large effective area;
randomly combining a plurality of groups of double-peak Brillouin scattering spectrums to form a Brillouin scattering spectrum image matrix, wherein the Brillouin scattering spectrum image matrix is used as a training set of a convolutional neural network;
training the convolutional neural network by adopting a training set, and iteratively optimizing parameters of the convolutional neural network by utilizing an error back propagation algorithm to obtain the convolutional neural network after training;
and (3) taking actually measured brillouin scattering spectrum data as a test set, inputting the test set into a convolutional neural network after training, and simultaneously obtaining temperature and strain information at the output end of the network.
And obtaining the bimodal Brillouin scattering spectrum in a numerical simulation mode.
The expression formula of the bimodal brillouin scattering spectrum is:
where g (v) denotes the normalized bimodal brillouin spectrum, v denotes the frequency of the incident light,and->Gains of two scattering spectrum peaks are respectively represented; />And->Respectively representing the brillouin frequency shift corresponding to two scattering spectrum peaks, ">And->Respectively representing the spectral widths of the two scatter spectra.
Optionally, the frequency corresponding to each brillouin scattering spectrum peak, i.e. the brillouin frequency shift, is linear with the amount of change in temperature and strain.
On the other hand, the system for simultaneously extracting the temperature and strain information in the Brillouin scattering spectrum image comprises a data acquisition module, a training set construction module, a model training module and a data processing module which are connected in sequence; wherein,
the data acquisition module is used for acquiring a double-peak Brillouin scattering spectrum of the optical fiber with a large effective area and a Brillouin frequency shift temperature coefficient and a strain coefficient corresponding to a spectrum peak;
the training set construction module is used for randomly combining a plurality of groups of the bimodal brillouin scattering spectrums to form a brillouin scattering spectrum image matrix, and the brillouin scattering spectrum image matrix is used as a training set of the convolutional neural network;
the model training module is used for training the convolutional neural network by adopting the training set, and iteratively optimizing parameters of the convolutional neural network by utilizing an error back propagation algorithm to obtain a trained convolutional neural network;
the data processing module is used for taking actually measured brillouin scattering spectrum data as a test set, inputting the actually measured brillouin scattering spectrum data into the convolutional neural network after training is completed, and simultaneously obtaining temperature and strain information at the output end of the network.
Compared with the prior art, the invention discloses a method and a system for simultaneously extracting temperature and strain information in a brillouin scattering spectrum image, which have the following beneficial technical effects:
(1) The technical scheme for solving the problem of cross sensitivity of temperature and strain measurement in the prior art is to process one-dimensional data of the brillouin scattering spectrum, such as fitting and the like, and the invention regards the brillouin scattering spectrum data obtained by adopting large-effective-area optical fiber measurement as a two-dimensional image, processes the two-dimensional image, and is a new thought for solving the problem of high-efficiency extraction of temperature and strain information at the same time;
(2) The method for processing the brillouin scattering spectrum image data by adopting the convolutional neural network does not need to adopt a time-consuming curve fitting method to determine the brillouin frequency shift or process the multimodal brillouin scattering spectrum point by point, can process all the brillouin scattering data along the sensing optical fiber at one time and extract temperature and strain information at the same time, and improves the system measurement speed while ensuring the measurement precision; the method can realize high-precision and simultaneous extraction of temperature and strain information, solves the problem of cross sensitivity of the Brillouin optical fiber sensing system, and greatly shortens the data processing time, thereby improving the system performance and meeting the requirements of application occasions requiring simultaneous measurement of temperature and strain.
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 a flow chart of the method of the present invention;
FIG. 2 is a block diagram of a convolutional neural network model of the present invention;
FIG. 3 is a flow chart of convolutional neural network model training of the present invention;
fig. 4 is a system configuration diagram of the present invention.
Detailed Description
The following description of the embodiments of the present invention 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 invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiment of the invention discloses a method for simultaneously extracting temperature and strain information in a brillouin scattering spectrum image, which is shown in fig. 1, and specifically comprises the following steps:
s1, acquiring a double-peak Brillouin scattering spectrum of an optical fiber with a large effective area, and a Brillouin frequency shift temperature coefficient and a strain coefficient corresponding to a spectrum peak;
s2, randomly combining a plurality of groups of double-peak Brillouin scattering spectrums to form a Brillouin scattering spectrum image matrix, and taking the Brillouin scattering spectrum image matrix as a training set of a convolutional neural network;
s3, training the convolutional neural network by adopting a training set, and iteratively optimizing parameters of the convolutional neural network by utilizing an error back propagation algorithm to obtain the convolutional neural network after training;
s4, taking actually measured brillouin scattering spectrum data as a test set, inputting the actually measured brillouin scattering spectrum data into a convolutional neural network after training, and simultaneously obtaining temperature and strain information at the output end of the network.
In this embodiment, the steps specifically include:
1) And obtaining the Brillouin scattering spectrum of the optical fiber with large effective area in a numerical simulation mode, wherein the scattering spectrum has two Brillouin scattering spectrum peaks, and parameters such as Brillouin frequency shift, spectrum width, signal to noise ratio and the like in each scattering spectrum are different. Wherein, the bimodal brillouin scattering spectrum is expressed by the following formula:
wherein,the value is 1; />0.15-0.55 can be taken at intervals of 0.1; />30-60MHz can be taken at intervals of 3 MHz; />40-75MHz can be taken at intervals of 5MHz; the temperature is 30-70 ℃ at intervals of 2 ℃; the strain was taken at intervals of 60. Mu.. Epsilon.to 1800. Mu.. Epsilon.with a spacing of 0. Mu.. Epsilon.to 1800. Mu.epsilon.
The brillouin frequency shift temperature coefficient and the brillouin frequency shift strain coefficient corresponding to each brillouin scattering spectrum peak need to be calibrated in advance. Through experimental calibration, the Brillouin frequency shift temperature coefficient of the first spectrum peakThe value is 1.12 MHz/DEG C, and the Brillouin frequency shift strain coefficient is +.>The value is 0.0385 MHz/mu epsilon; brillouin frequency shift temperature coefficient of the second spectral peak>The value is 0.96 MHz/DEG C, and the Brillouin frequency shift strain coefficient is +.>The value is 0.0383 MHz/mu epsilon.
2) And respectively selecting 1000 groups of brillouin scattering spectrums generated by the steps to randomly generate an image matrix. Since the sweep frequency range is 10.598 GHz-10.964 GHz and the interval is 2MHz in the experimental process, the size of each image matrix is 1000×184.
3) A convolutional neural network model is constructed, the structure of which is shown in fig. 2.
The size of an input layer of the convolutional neural network is 1000 multiplied by 184, input data firstly passes through a convolutional layer with 64 3 multiplied by 3 convolutional kernels to obtain 64 feature graphs, and then the dimension of the feature graphs is reduced through the maximum pooling operation; then, inputting the characteristic diagram into a deep sub-network constructed based on the residual network principle, wherein the sub-network structure is 1×1-3×3-1×1, namely, the sub-network structure comprises 3 convolution layers, the corresponding convolution kernel sizes of the sub-network structure are 1×1, 3×3 and 1×1 respectively, and repeating the sub-network structure for 3 times to form 9 layers; and finally, obtaining the output containing temperature and strain information through 7 layers of common deep networks. The residual network structure can solve the performance degradation problem of the deep network, so that the network expression capacity is better, and the network training speed is increased.
4) And (3) iteratively optimizing parameters of the convolutional neural network by using an error back propagation algorithm to finish training of the neural network, wherein the training process is shown in figure 3.
Because each group of brillouin scattering spectrum data corresponds to a group of temperature and strain values, the convolutional neural network obtained through effective training realizes the mapping of the brillouin scattering spectrum data and the temperature and strain information. For the convolutional neural network, if the number of scattering spectrums in the brillouin scattering spectrum image is represented by N, and M represents the number of sampling points of each scattering spectrum, the network input data is an n×m brillouin scattering spectrum image matrix, and the output data is an n×2 temperature and strain value matrix.
5) And the Brillouin Optical Time Domain Reflection (BOTDR) or the Brillouin Optical Time Domain Analysis (BOTDA) system is used for measuring the brillouin scattering spectrum data of the optical fiber with large effective area as the input of the convolutional neural network, so that the temperature and strain information along the optical fiber can be obtained at the output end of the network at the same time, and the simultaneous measurement of the temperature and strain information along the optical fiber is realized.
Analysis of the results: and selecting an image matrix formed by 1000 groups of brillouin scattering spectrum data as input of a convolutional neural network, and analyzing extraction results of temperature and strain information. The errors of the temperature and strain extraction results (root mean square error RMSE and standard deviation SD) are shown in table 1 when the brillouin spectral signal-to-noise ratio is 26 dB.
TABLE 1
Therefore, compared with a method for processing one-dimensional data, the convolutional neural network can process a plurality of groups of brillouin scattering spectrum data along the sensing optical fiber at one time by setting the size of the network input matrix, the extraction precision of temperature and strain information is better than that of an equation solving method, the extraction time of the sensing information is greatly shortened, the system measurement time is effectively shortened, and the performance of the brillouin optical fiber sensing system is improved.
Further, for a more intuitive and detailed description of technical effects, the following is now stated: the brillouin shift is linearly dependent on the amount of change in temperature and strain, and can be expressed as
v B (T,ε)=v B (T 0 ,ε 0 )+C vT (T-T 0 )+C vε (ε-ε 0 )=v B0 +C vT ΔT+C vε Δε
Wherein T is 0 And epsilon 0 Respectively the initial temperature and strain, v B0 Representing the brillouin shift under initial temperature and strain conditions, C vT Temperature coefficient representing brillouin frequency shift, C vε The strain coefficient indicating the brillouin shift, and Δt and Δε indicate the amounts of change in temperature and strain, respectively. Therefore, when the temperature and strain at a certain position of the optical fiber are changed at the same time, the two cannot be distinguished by a single brillouin shift.
Taking a large effective area optical fiber as an example, a brillouin scattering spectrum of the optical fiber has a plurality of scattering peaks, and the brillouin frequency shift variation corresponding to two brillouin scattering peaks in the brillouin scattering spectrum of the optical fiber with the large effective area is measured as follows:
in ΔBFS 1 And ΔBFS 2 Respectively representing the variation of Brillouin frequency shift corresponding to two spectral peaks, C vT1 And C vT2 Respectively represent the temperature coefficients of Brillouin frequency shift corresponding to two spectral peaks, C vε1 And C vε2 Respectively represent the brillouin frequency shift strain coefficients corresponding to the two spectral peaks. After finishing, the variation of temperature and strain is obtained
Because the brillouin frequency shift temperatures and the strain coefficients corresponding to two brillouin scattering peaks in the same sensing optical fiber are very close, larger errors are easily generated when solving the equation set, and the measurement accuracy of temperature and strain is reduced. In addition, if the brillouin frequency shift is extracted by adopting a curve fitting method, longer data processing time is required, and the system measurement speed is affected. Therefore, the method and the device can avoid the problem that the time-consuming process of determining the Brillouin frequency shift by curve fitting and solving the equation set are prone to generating larger errors.
The embodiment 2 of the invention discloses a system for simultaneously extracting temperature and strain information in a brillouin scattering spectrum image, which is shown in fig. 4 and comprises a data acquisition module, a training set construction module, a model training module and a data processing module which are connected in sequence; wherein,
the data acquisition module is used for acquiring the double-peak Brillouin scattering spectrum of the optical fiber with large effective area and the Brillouin frequency shift temperature coefficient and the strain coefficient corresponding to the spectral peak;
the training set construction module is used for randomly combining a plurality of groups of bimodal Brillouin scattering spectra to form a Brillouin scattering spectrum image matrix, and the Brillouin scattering spectrum image matrix is used as a training set of the convolutional neural network;
the model training module is used for training the convolutional neural network by adopting a training set, and iteratively optimizing parameters of the convolutional neural network by utilizing an error back propagation algorithm to obtain the convolutional neural network after the training is completed;
the data processing module is used for taking the actually measured brillouin scattering spectrum data as a test set, inputting the test set into the convolutional neural network after training, and simultaneously obtaining temperature and strain information at the output end of the network.
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. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. 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 invention. Thus, the present invention 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 (5)
1. A method for simultaneously extracting temperature and strain information in a Brillouin scattering spectrum image is characterized by comprising the following specific steps:
acquiring a brillouin frequency shift temperature coefficient and a strain coefficient corresponding to a double-peak brillouin scattering spectrum of the optical fiber with a large effective area;
randomly combining a plurality of groups of the bimodal brillouin scattering spectrums to form a brillouin scattering spectrum image matrix, wherein the brillouin scattering spectrum image matrix is used as a training set of a convolutional neural network;
training the convolutional neural network by adopting the training set, and iteratively optimizing parameters of the convolutional neural network by utilizing an error back propagation algorithm to obtain a trained convolutional neural network; n represents the number of scattering spectrums in the Brillouin scattering spectrum image, M represents the number of sampling points of each scattering spectrum, the network input data is an N multiplied by M Brillouin scattering spectrum image matrix, and the output data is an N multiplied by 2 temperature and strain value matrix;
and taking actually measured brillouin scattering spectrum data as a test set, inputting the test set into the convolutional neural network after training, and simultaneously obtaining temperature and strain information at the network output end.
2. The method for simultaneously extracting temperature and strain information in a brillouin spectrum image according to claim 1, wherein the bimodal brillouin spectrum is obtained by means of numerical simulation.
3. The method for simultaneously extracting temperature and strain information in a brillouin scattering spectrum image according to claim 1, wherein the expression formula of the bimodal brillouin scattering spectrum is:
where g (v) denotes the normalized bimodal brillouin spectrum, v denotes the frequency of the incident light,and->Gains of two scattering spectrum peaks are respectively represented; />And->Respectively representing the brillouin frequency shift corresponding to two scattering spectrum peaks, ">Andrespectively representing the spectral widths of the two scatter spectra.
4. The method for simultaneously extracting temperature and strain information in a brillouin scattering spectrum image according to claim 1, wherein the frequency corresponding to each brillouin scattering spectrum peak, i.e. brillouin frequency shift, is respectively in linear relation with the temperature variation and the strain variation.
5. The system for simultaneously extracting the temperature and strain information in the Brillouin scattering spectrum image is characterized by comprising a data acquisition module, a training set construction module, a model training module and a data processing module which are connected in sequence; wherein,
the data acquisition module is used for acquiring a double-peak Brillouin scattering spectrum of the optical fiber with a large effective area and a Brillouin frequency shift temperature coefficient and a strain coefficient corresponding to a spectrum peak;
the training set construction module is used for randomly combining a plurality of groups of the bimodal brillouin scattering spectrums to form a brillouin scattering spectrum image matrix, and the brillouin scattering spectrum image matrix is used as a training set of the convolutional neural network;
the model training module is used for training the convolutional neural network by adopting the training set, and iteratively optimizing parameters of the convolutional neural network by utilizing an error back propagation algorithm to obtain a trained convolutional neural network; n represents the number of scattering spectrums in the Brillouin scattering spectrum image, M represents the number of sampling points of each scattering spectrum, the network input data is an N multiplied by M Brillouin scattering spectrum image matrix, and the output data is an N multiplied by 2 temperature and strain value matrix;
the data processing module is used for taking actually measured brillouin scattering spectrum data as a test set, inputting the actually measured brillouin scattering spectrum data into the convolutional neural network after training is completed, and simultaneously obtaining temperature and strain information at the output end of the network.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210203784.2A CN114518182B (en) | 2022-03-02 | 2022-03-02 | Method and system for simultaneously extracting temperature and strain information in brillouin scattering spectrum image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210203784.2A CN114518182B (en) | 2022-03-02 | 2022-03-02 | Method and system for simultaneously extracting temperature and strain information in brillouin scattering spectrum image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114518182A CN114518182A (en) | 2022-05-20 |
CN114518182B true CN114518182B (en) | 2024-03-22 |
Family
ID=81598481
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210203784.2A Active CN114518182B (en) | 2022-03-02 | 2022-03-02 | Method and system for simultaneously extracting temperature and strain information in brillouin scattering spectrum image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114518182B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105181152A (en) * | 2015-08-28 | 2015-12-23 | 苏州光格设备有限公司 | Calculation method for frequency shift of distributed Brillouin scattered spectrum |
CN107402082A (en) * | 2016-05-20 | 2017-11-28 | 北京自动化控制设备研究所 | A kind of brillouin scattering signal processing method and its distributed optical fiber sensing system |
CN110926355A (en) * | 2019-11-07 | 2020-03-27 | 华中科技大学 | Brillouin frequency shift extraction method and device based on convolutional neural network |
CN112781634A (en) * | 2021-04-12 | 2021-05-11 | 南京信息工程大学 | BOTDR distributed optical fiber sensing system based on YOLOv4 convolutional neural network |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10540590B2 (en) * | 2016-12-29 | 2020-01-21 | Zhejiang Gongshang University | Method for generating spatial-temporally consistent depth map sequences based on convolution neural networks |
US10713563B2 (en) * | 2017-11-27 | 2020-07-14 | Technische Universiteit Eindhoven | Object recognition using a convolutional neural network trained by principal component analysis and repeated spectral clustering |
-
2022
- 2022-03-02 CN CN202210203784.2A patent/CN114518182B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105181152A (en) * | 2015-08-28 | 2015-12-23 | 苏州光格设备有限公司 | Calculation method for frequency shift of distributed Brillouin scattered spectrum |
CN107402082A (en) * | 2016-05-20 | 2017-11-28 | 北京自动化控制设备研究所 | A kind of brillouin scattering signal processing method and its distributed optical fiber sensing system |
CN110926355A (en) * | 2019-11-07 | 2020-03-27 | 华中科技大学 | Brillouin frequency shift extraction method and device based on convolutional neural network |
CN112781634A (en) * | 2021-04-12 | 2021-05-11 | 南京信息工程大学 | BOTDR distributed optical fiber sensing system based on YOLOv4 convolutional neural network |
Also Published As
Publication number | Publication date |
---|---|
CN114518182A (en) | 2022-05-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108955734B (en) | Cavity length demodulation method of optical fiber F-P temperature/pressure composite sensor | |
CN107490397B (en) | High-accuracy self-adaptation filters the quick Peak Search Method of FBG spectrum | |
WO2020034558A1 (en) | High-precision temperature demodulation method oriented toward distributed fiber raman sensor | |
CN106017522B (en) | A kind of quick high accuracy signal demodulating method of fibre-optical F-P sensor | |
CN103674864B (en) | A kind of fish water content distribution detection method based on high light spectrum image-forming technology | |
CN108534811B (en) | Cavity length demodulation algorithm of short-cavity optical fiber Fabry-Perot sensor | |
CN102426198A (en) | Acoustic emission signal sensing system based on matching-type fiber Bragg grating (FBG) | |
CN108562237B (en) | Device and method for performing spectrum calibration in optical frequency domain reflection sensing system by adopting HCN (hydrogen cyanide) air chamber | |
CN113866102B (en) | Soil health investigation monitoring method based on spectrum | |
CN103487074A (en) | Method for processing FBG (fiber bragg grating) sensing signal by utilizing three-point peek-seeking algorithm | |
CN105841723A (en) | FBG sensor demodulation peak-searching method based on wavelet singularity detection | |
CN109282837B (en) | Demodulation method of Bragg fiber grating staggered spectrum based on LSTM network | |
CN101750024A (en) | Near infrared spectrum detection method for insulation coating layer thickness of silicon steel | |
CN104344890A (en) | Weak light signal spectrum fast test device and method | |
CN108225385A (en) | A kind of overlapping FBG transducing signal peak value localization methods | |
CN106643835A (en) | Fiber fabry-perot cavity demodulation method, device thereof and fiber fabry-perot interferometer | |
CN114518182B (en) | Method and system for simultaneously extracting temperature and strain information in brillouin scattering spectrum image | |
CN101236074A (en) | Method for measuring strain distribution using optical fier grating | |
CN116105776A (en) | Method and device for high-precision in-situ calibration and high-speed real-time demodulation of cavity length of optical fiber Fabry-Perot sensor | |
CN113465676B (en) | Double-stage Brillouin frequency shift extraction method | |
Harasim et al. | Accuracy improvement of Fiber Bragg Grating peak wavelength demodulation using wavelet transform and various center wavelength detection algorithms | |
CN110672058B (en) | On-line calibration data sequence matching method and device of sensor for structure monitoring | |
CN106872397A (en) | A kind of method based on existing calibration model quick detection agricultural product chemical constituent | |
Yu et al. | Influence of light intensity distribution characteristics of light source on measurement results of canopy reflectance spectrometers | |
CN111707303A (en) | Variable-step-length rapid high-precision signal demodulation method for optical fiber F-P sensor |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |