CN117109720B - Distributed optical fiber vibration sensor based on time division multiplexing technology - Google Patents
Distributed optical fiber vibration sensor based on time division multiplexing technology Download PDFInfo
- Publication number
- CN117109720B CN117109720B CN202311132994.8A CN202311132994A CN117109720B CN 117109720 B CN117109720 B CN 117109720B CN 202311132994 A CN202311132994 A CN 202311132994A CN 117109720 B CN117109720 B CN 117109720B
- Authority
- CN
- China
- Prior art keywords
- vibration
- optical fiber
- sensor
- sensing
- result
- 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
- 239000013307 optical fiber Substances 0.000 title claims abstract description 139
- 238000005516 engineering process Methods 0.000 title claims abstract description 24
- 238000001514 detection method Methods 0.000 claims description 84
- 230000004927 fusion Effects 0.000 claims description 29
- 238000000253 optical time-domain reflectometry Methods 0.000 claims description 19
- 238000000034 method Methods 0.000 claims description 18
- 239000000835 fiber Substances 0.000 claims description 12
- 230000006698 induction Effects 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 10
- 230000004913 activation Effects 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 8
- 238000011176 pooling Methods 0.000 claims description 7
- 238000003909 pattern recognition Methods 0.000 claims description 5
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000010521 absorption reaction Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 3
- 238000012935 Averaging Methods 0.000 claims 1
- 230000004044 response Effects 0.000 description 14
- 238000002592 echocardiography Methods 0.000 description 13
- 230000006870 function Effects 0.000 description 10
- 230000003287 optical effect Effects 0.000 description 10
- 230000000694 effects Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 239000000523 sample Substances 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000001427 coherent effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008033 biological extinction Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 230000005428 wave function Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H9/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
- G01H9/004—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means using fibre optic sensors
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Abstract
The application discloses a distributed optical fiber vibration sensor based on a time division multiplexing technology, which relates to the field of distributed optical fiber acoustic wave sensing systems.
Description
Technical Field
The application relates to the field of distributed optical fiber acoustic wave sensing systems, in particular to a distributed optical fiber vibration sensor based on a time division multiplexing technology.
Background
The phi-OTDR (Phase-SENSITIVE OPTICAL TIME DOMAIN REFLECTOMETE, phase sensitive optical time domain reflection) is a sensing technology of DAS (Distributed Optical Fiber Audio SENSING SYSTEM, distributed optical fiber acoustic wave sensing system) through acousto-optic sensing, and an optical fiber vibration sensor based on the phi-OTDR utilizes a Rayleigh backscattering principle to realize sensing of external vibration signals, so that distributed continuous detection and positioning of vibration sources are performed.
The phi-OTDR system has the inherent defects of the distributed optical fiber acoustic wave sensing system in practical application, namely the problem that the sensing distance and the frequency response bandwidth are mutually restricted, and the frequency response bandwidth is reduced along with the increase of the sensing distance of the system, because the time interval of the detection pulse light injected into the sensing optical fiber is required to be longer than the round trip time of the detection pulse light in the sensing optical fiber, otherwise, the Rayleigh scattering echoes of continuous detection pulse light are overlapped and can not be distinguished, so that the longer the distance of the sensing optical fiber is, the more limited the repetition frequency of the detection pulse light is, and the frequency response bandwidth is influenced approximately. This drawback limits the application of distributed fiber optic acoustic wave sensing systems in the field of vibration signal detection.
Disclosure of Invention
The present inventors have proposed a distributed optical fiber vibration sensor based on a time division multiplexing technology, and the technical scheme of the present application is as follows:
A distributed optical fiber vibration sensor based on a time division multiplexing technology comprises a tunable laser, an electroabsorption modulator, a sensing channel and a control processor;
the control processor is connected with and controls the tunable laser, and the tunable laser outputs continuous laser light containing N wavelengths;
the electroabsorption modulator is used for modulating continuous laser generated by the tunable laser into detection pulse light containing N wavelengths and inputting the detection pulse light into the signal input end of the sensing channel;
The sensing channel comprises sensing optical fibers, N wavelength division demultiplexers, N circulators and photoelectric detectors, the sensing optical fibers are arranged in a vibration area to be detected, and the sensing optical fibers comprise N continuous optical fiber sections, wherein N is more than or equal to 2; in the sensing channel, the input end of a first wavelength division multiplexer is connected with the signal input end of the sensing channel, for any integer parameter 1N-1, the first output end of an N-th wavelength division multiplexer is connected with the input end of an n+1th wavelength division multiplexer through a delay optical fiber with the same length as that of an N-th optical fiber section, and the first output end of the N-th wavelength division multiplexer is suspended; the second output end of the nth wavelength division demultiplexer is connected with the first port of the nth circulator, the second port of the nth circulator is connected with the optical fiber incident end of the nth optical fiber section, and the optical fiber emergent end of the nth optical fiber section is connected with the third port of the (n+1) th circulator; the optical fiber emergent end of the Nth optical fiber section is suspended, the third port of the first circulator is connected with a photoelectric detector, and the signal output end of the photoelectric detector is connected with the signal output end of the sensing channel; each wavelength division multiplexer corresponds to one wavelength, each wavelength division multiplexer outputs detection pulse light with the corresponding wavelength through a second output end, outputs detection pulse light with other wavelengths through a first output end, the second output ends of the N wavelength division multiplexers respectively output N detection pulse light with different wavelengths, and signals in each circulator are sequentially transmitted according to a first port, a second port and a third port;
The signal output end of the sensing channel is connected with the control processor, and the control processor outputs a vibration induction result based on the phi-OTDR technology according to the vibration detection signal output by the photoelectric detector, wherein the vibration induction result is used for indicating the vibration disturbance suffered by the sensing optical fiber in the sensing channel.
The distributed optical fiber vibration sensor comprises an optical fiber amplifier and M sensing channels, wherein sensing optical fibers in each sensing channel are respectively distributed at different spatial positions of a region to be detected for vibration, and integer parameter M is more than or equal to 2;
The input end of the optical fiber amplifier is connected with the output end of the electroabsorption modulator, each output end of a 1*M coupler contained in the optical fiber amplifier is respectively connected with the signal input ends of M transmission channels, the signal output ends of the transmission channels are all connected with the control processor, and the control processor synthesizes vibration detection signals output by the M photoelectric detectors and outputs vibration induction results based on the phi-OTDR technology.
The further technical scheme is that the vibration detection signals output by the M photoelectric detectors are synthesized to output vibration induction results based on the phi-OTDR technology, and the method comprises the following steps:
obtaining an inductor result based on the phi-OTDR technology according to the vibration detection signal output by each photoelectric detector, wherein the inductor result is used for indicating whether the target position in the area to be detected for vibration has vibration or not;
Building an identification framework, wherein the identification framework comprises 4 mutually exclusive propositions respectively: the target position is actually vibrated and the sensor result indicates that there is vibration, the target position is actually vibrated and the sensor result indicates that there is no vibration, the target position is actually non-vibrated and the sensor result indicates that there is vibration, the target position is actually non-vibration and the sensor result indicates that there is no vibration;
Initializing the basic probability that each sensor result corresponds to each proposition under the identification frame, and carrying out information fusion on M sensor results according to the identification frame based on the D-S evidence theory to obtain a vibration sensing result, wherein the vibration sensing result is used for indicating whether the target position has vibration or not.
According to a further technical scheme, the method for obtaining the vibration induction result by information fusion of M inductor results according to the identification frame based on the D-S evidence theory comprises the following steps:
According to the similarity between any two inductor results in the M inductor results, calculating to obtain the credibility weight of each inductor result, and carrying out weighted average on the basic probability of each inductor result corresponding to the proposition according to the credibility weight of each inductor result to obtain the weighted average probability of each proposition;
And carrying out information fusion on the M sensor results according to the identification frame based on the D-S evidence theory according to the weighted average probability of each proposition to obtain a vibration sensing result.
The further technical scheme is that the obtaining of the weighted average probability of each proposition comprises the following steps:
Determining the weight of any i-th proposition A i M j(Ai) is the basic probability corresponding to the ith proposition A i in the jth sensor result, the integer parameter 1 is less than or equal to j is less than or equal to M, and the integer parameter 1 is less than or equal to i is less than or equal to 4;
Calculating the similarity of any j-th sensor result m j and g-th sensor result m g by combining the weight of each proposition M g(Ai) is the basic probability corresponding to the ith proposition A i in the g-th inductor result, and the integer parameter is 1-g-M;
Calculating the weight of any jth sensor result And carrying out normalization processing to obtain the credibility weight/>, of the j-th sensor result
Each sensor result is weighted and averaged according to the credibility weight of each sensor result, and the weighted average probability of any ith proposition A i is determined
The further technical scheme is that initializing the basic probability that each sensor result corresponds to each proposition under the identification frame comprises the following steps of for any jth sensor result:
When the j-th sensor result indicates that the target position has vibration, determining that m j(A2)＝mj(A4) =0, determining m j(A1) and m j(A3)＝1-mj(A1 according to the real-time signal-to-noise ratio), wherein the larger the signal-to-noise ratio is, the larger the value of m p(A1) is;
When the j-th sensor result indicates that the target position has no vibration, determining that m j(A1)＝mj(A3) =0, determining m j(A4) and m j(A2)＝1-mj(A4 according to the real-time signal-to-noise ratio), wherein the larger the signal-to-noise ratio is, the larger the value of m j(A4) is;
Wherein, proposition A 1 indicates that the target position is actually vibrated and the sensor result indicates that there is vibration, proposition A 2 indicates that the target position is actually vibrated and the sensor result indicates that there is no vibration, proposition A 3 indicates that the target position is actually not vibrated and the sensor result indicates that there is vibration, proposition A 4 indicates that the target position is actually not vibrated and the sensor result indicates that there is no vibration, and integer parameter 1 is less than or equal to j is less than or equal to M.
The further technical proposal is that when the jth sensor result indicates that the target position has vibration, the method comprises the following steps ofReal-time signal to noise ratio/>Mapping to a value interval of (0.5, 1) to obtain m j(A1);
When the jth sensor result indicates that the target position has no vibration, the method is as follows Real-time signal to noise ratio/>M j(A4 is obtained in the value interval mapped to (0.8,1).
The further technical scheme is that the information fusion of the M inductor results to obtain the vibration induction result further comprises:
When the vibration sensing result is determined to be used for indicating that the target position vibrates, respectively inputting the vibration detection signals output by each photoelectric detector into a vibration mode identification model to obtain the vibration modes of each vibration detection signal output by the vibration mode identification model, voting the vibration modes of each vibration detection signal by adopting a voting mechanism to determine a final vibration mode, and then the obtained vibration sensing result is also used for indicating the final vibration mode;
The vibration mode identification model is obtained in advance based on end-to-end convolutional neural network training.
The further technical scheme is that the vibration mode identification model sequentially comprises the following components from input to output: the system comprises a plurality of feature extraction layers, an unfolding layer, a full-connection layer and a classifier which are stacked, wherein each feature extraction layer sequentially comprises a convolution layer, an activation function and a pooling layer.
The further technical scheme is that the activation function adopts a ReLU and the classifier adopts a softmax.
The beneficial technical effects of the application are as follows:
The application discloses a distributed optical fiber vibration sensor based on a time division multiplexing technology, which outputs continuous laser with a plurality of wavelengths through a tunable laser, modulates the continuous laser into detection light pulses with a plurality of wavelengths through an electroabsorption modulator, and each wavelength division multiplexer in a sensing channel respectively demodulates the detection light pulse with a corresponding wavelength to be injected into a corresponding sensing optical fiber, so that optical signals on different optical fiber sections sequentially reach a photoelectric detector to form time division multiplexing through the wavelength division multiplexing technology, thereby leading the detection pulse light with a plurality of wavelengths to exist in the whole sensing optical fiber, and improving the pulse repetition frequency to be multiple times of the original under the condition that the sensing optical fiber has the same length, and leading the frequency response bandwidth to be also improved to be multiple times of the original, so that the distributed optical fiber vibration sensor can give consideration to the detection distance and the detection frequency range.
Furthermore, on the basis that a single sensing channel adopts the time division multiplexing structure, a plurality of sensing channels can be expanded and used, and a final vibration sensing result is obtained by carrying out information fusion on multiple paths of sensing signals, so that the anti-interference capability, the stability of a system and the reliability and fault tolerance of an identification result of the whole distributed optical fiber vibration sensor are greatly improved.
And when information fusion is carried out, the improved D-S evidence theory is adopted to carry out multi-sensing information fusion, the sensor results of each sensing channel are subjected to data pre-modification in advance, and the accuracy and feasibility of the results are improved. And when vibration is detected, a second-stage vibration detection can be added, and the end-to-end convolutional neural network is utilized to identify the vibration mode, so that the type of the alarm signal is accurately judged.
Drawings
FIG. 1 is a schematic diagram of a distributed fiber optic vibration sensor according to one embodiment of the present application.
Fig. 2 is a schematic structural view of a distributed optical fiber vibration sensor according to another embodiment of the present application.
FIG. 3 is a network architecture diagram of a vibration pattern recognition model according to one embodiment of the present application.
Detailed Description
The following describes the embodiments of the present application further with reference to the drawings.
The application discloses a distributed optical fiber vibration sensor based on a time division multiplexing technology, referring to a structural diagram shown in fig. 1, the distributed optical fiber vibration sensor comprises a tunable laser, an electroabsorption modulator, a sensing channel and a control processor.
Each sensing channel comprises a sensing optical fiber, N wavelength division demultiplexers, N circulators and a photoelectric detector. The sensing optical fiber is arranged in a vibration area to be detected, and comprises N continuous optical fiber sections, wherein N is more than or equal to 2. As shown in fig. 1, the first to nth fiber segments are denoted as f_1 to f_n, respectively. The first to nth circulators are respectively marked as c_1 to c_n, and signals in each circulator are sequentially transmitted according to the first port, the second port and the third port. The first to nth demultiplexers are denoted as demultiplexers_1 to_n, respectively.
In the sensing channel, the input end of the first wavelength division multiplexer is connected with the signal input end of the sensing channel, and for any integer parameter 1N is less than or equal to N-1, the first output end of the nth wavelength division multiplexer is connected with the input end of the (n+1) th wavelength division multiplexer through a delay optical fiber with the same length as the nth optical fiber section, so that the sensing channel actually further comprises N-1 delay optical fibers which are respectively marked as FD_1 to FD_N-1. The first output end of the Nth wavelength division multiplexer is suspended.
The second output end of any nth wavelength division demultiplexer is connected with the first port of the nth circulator, the second port of the nth circulator is connected with the optical fiber incident end of the nth optical fiber section, and the optical fiber emergent end of the nth optical fiber section is connected with the third port of the (n+1) th circulator. The optical fiber emergent end of the Nth optical fiber section is suspended. The third port of the first circulator is connected with a photoelectric detector, and the signal output end of the photoelectric detector is connected with the signal output end of the sensing channel.
The connection between the tunable laser, the electro-absorption modulator, the sensing channel and the control processor includes: the tunable laser is connected with the input end of the electroabsorption modulator, the output end of the electroabsorption modulator is connected with the signal input end of the sensing channel, the signal output end of the sensing channel is connected with the control processor,
The control processor is also connected to control the tunable laser so that the tunable laser outputs continuous laser light containing N wavelengths, and the electroabsorption modulator is used for modulating the continuous laser light generated by the tunable laser into detection pulse light containing N wavelengths (the wavelengths are respectively denoted as λ1 and λ2 … … λn) and inputting the detection pulse light into the signal input end of the sensing channel. Each wavelength division multiplexer in the transmission channel corresponds to one wavelength, each wavelength division multiplexer outputs detection pulse light with the corresponding wavelength through the second output end, and outputs detection pulse light with other wavelengths through the first output end, and the second output ends of the N wavelength division multiplexers respectively output detection pulse light with N different wavelengths. Therefore, assuming that the first to nth demultiplexers correspond to wavelengths λ1 to λn, respectively, the probe pulse light containing N wavelengths is input to the sensing channel:
After the detection pulse light with N wavelengths passes through the first demultiplexer, the second output end of the first demultiplexer outputs the detection pulse light with wavelength lambda 1, and the detection pulse light with other wavelengths is output through the first output end of the first demultiplexer.
The detection pulse light with the wavelength lambda 1 output by the second output end of the first wavelength division multiplexer enters the first port of the first circulator, is sequentially transmitted to the second port of the first circulator and then enters the optical fiber incident end of the first optical fiber section. When the detection pulse light with the wavelength lambda 1 is transmitted in the first optical fiber section, backward Rayleigh scattering echoes are continuously generated at the optical fiber incident end of the first optical fiber section and enter the second port of the first circulator, and the detection pulse light is sequentially transmitted to the third port of the first circulator and then enters the photoelectric detector.
Meanwhile, the detection pulse light with other wavelengths output by the first output end of the first wavelength division multiplexer continues to be transmitted forwards to enter the second wavelength division multiplexer, after passing through the second wavelength division multiplexer, the second output end of the second wavelength division multiplexer outputs the detection pulse light with the wavelength lambda 2, and the detection pulse light with other wavelengths continues to be transmitted forwards through the first output end of the second wavelength division multiplexer.
The detection pulse light with the wavelength lambda 2 output by the second output end of the second wavelength division multiplexer enters the first port of the second circulator, sequentially transmits to the second port of the second circulator and then enters the optical fiber incidence end of the second optical fiber section, and when the detection pulse light with the wavelength lambda 2 is transmitted in the second optical fiber section, backward Rayleigh scattering echoes are continuously generated at the optical fiber incidence end of the second optical fiber section and enter the second port of the second circulator, sequentially transmits to the third port of the second circulator and then enters the first optical fiber section, sequentially transmits to the third port of the first circulator and then enters the photoelectric detector.
Meanwhile, the detection pulse light with other wavelengths output by the first output end of the second wavelength division multiplexer continues to be transmitted forwards to enter the third wavelength division multiplexer, and the above processes are repeated to be transmitted sequentially. Finally, backward Rayleigh scattering echoes in each optical fiber section sequentially reach the photoelectric detector according to the sequence of the wavelengths lambda 1 and lambda 2 … … lambda N to form a time division multiplexing system, so that N wavelength detection pulse lights exist in the whole sensing optical fiber, and the pulse repetition frequency can be increased to be N times under the condition that the sensing optical fiber has the same length, and the frequency response bandwidth is also increased to be N times.
The photoelectric detector in the sensing channel can receive backward Rayleigh scattering echoes of various wavelengths, and convert the backward Rayleigh scattering echoes into vibration detection signals to be output, and the control processor outputs vibration induction results based on the phi-OTDR technology according to the vibration detection signals output by the photoelectric detector, wherein the vibration induction results are used for indicating vibration disturbance to the sensing optical fibers in the sensing channel. The phi-OTDR technology based on backward Rayleigh scattered light is introduced as follows:
The light is transmitted in the sensing optical fiber, and various scattering phenomena are caused by the non-uniformity in the sensing optical fiber material, the Rayleigh scattering is elastic scattering, the incident light and the scattered light have the same frequency, the phi-OTDR technology mainly enables detection pulse light with higher power to be directly incident into the sensing optical fiber through the optical fiber incident end, the optical power of Rayleigh scattering echo axially and backwardly transmitted along the sensing optical fiber is detected at the optical fiber incident end, and the light power of Rayleigh scattering echo at the optical fiber incident end is detected by using a photoelectric detector because the scattered light power in the sensing optical fiber is proportional to the optical power of an incident point, so that the transmission information along the sensing optical fiber can be obtained.
However, the traditional phi-OTDR technology has weaker sensitivity to disturbance events, so that key parameters of vibration signals are obtained through phase detection of Rayleigh scattering echoes based on a coherent detection theory. The light source of the phi-OTDR is a narrow linewidth laser, and the sensing function is realized by detecting interference signals of Rayleigh scattering echoes among scattering points in the light pulse width. When the detection pulse light propagates in the sensing optical fiber, the refractive index distribution in the sensing optical fiber is uneven, and scattering occurs in each position, which is equivalent to the existence of a plurality of independent scattering units in the sensing optical fiber. Assuming that K scattering points exist in the sensing optical fiber in total, at time t, the optical fiber incident end of the sensing optical fiber receives rayleigh scattering echoes from the p-th to the q-th (q > p) scattering points, the rayleigh scattering echoes from the sensor optical fiber along the line show linear superposition of wave functions, and at this time, the total rayleigh scattering echo received by the optical fiber incident end of the sensing optical fiber can be expressed as:
Wherein E 0 is the amplitude of the detection pulse light incident from the incident end of the optical fiber, alpha is the attenuation coefficient of the optical fiber, z p is the distance from the p-th scattering point in the sensing optical fiber to the incident end of the optical fiber, r k is the scattering coefficient of the k-th scattering point, Is the scattered light phase of the kth scattering point and i is the imaginary unit.
When the light source meets the interference condition of backward Rayleigh scattered light, rayleigh scattered echo with the optical path difference smaller than the coherence length of the light source can interfere at the incidence end of the optical fiber, and the light intensity is as follows:
from the above equation, the intensity of the coherent rayleigh scattered light is related to the phase of all scattered light within the pulse width. When vibration disturbance occurs to the outside, the sensing optical fiber can generate tiny mechanical strain, and the length, the refractive index and the diameter of the sensing optical fiber can be changed, so that the phase information of light is affected, and the total phase delay is:
Wherein a is the diameter of the sensing optical fiber, and three terms on the right side of the above equal sign respectively represent stress effect caused by deformation of the sensing optical fiber, photoelastic effect caused by refractive index and poisson effect caused by diameter change. Therefore, the Rayleigh scattering echo can be divided into two parts, wherein one part of scattered light E A comes from a scattering point between a disturbance point and the incident end of the optical fiber, is not influenced by the disturbance, and has unchanged optical phase. Another portion of the scattered light E B comes from the scattering point between the disturbance point and the end of the fiber, and the optical phase adds an additional phase due to the perceived vibration disturbance. The two parts of scattered light are respectively:
Wherein the vibration disturbance occurs at the h scattering point of the sensing fiber, The light intensity of the Rayleigh scattering echo is as follows:
When the vibration disturbance occurs, only the Rayleigh scattering echo at the vibration disturbance position changes in light intensity, so that by tracking the change, whether the vibration disturbance occurs or not and the position of the vibration disturbance can be detected.
The main performance indexes of the distributed optical fiber vibration sensor of the application are as follows: detection length, spatial resolution, frequency response range, and positioning accuracy:
(1) The detection length is the distance which can be detected furthest after the DAS system is connected with the sensing optical fiber, and is mainly related to the scanning frequency f c of the electroabsorption modulator. The electroabsorption modulator is used for carrying out chopper modulation on continuous laser emitted by the tunable laser to make the continuous laser become detection pulse light, and then the expression of the detection length is that C is the propagation speed of light in vacuum and n is the refractive index of the sensing fiber.
(2) The positioning accuracy is mainly determined by the sampling frequency f s of the data acquisition card in the control processor. The positioning accuracy can be expressed by the length of the time backward Rayleigh scattering echo of one sampling period, which is walked in the sensing optical fiber, asIt can be seen that the higher the sampling frequency f s of the data acquisition card is, the greaterThe smaller the value, the higher the positioning accuracy.
(3) The spatial resolution is the minimum length that the distributed fiber vibration sensor can distinguish between two independent events. In general, it is controlled mainly by the optical pulse width T W of the probe pulse light injected into the sensing fiber, and can be expressed asIt can be seen that the larger the optical pulse width T W of the probe pulse light is, the greater the spatial resolution is.
(4) The frequency response range reflects the characterization of the external interference response frequency range by the distributed optical fiber vibration sensor, and the larger the frequency response range is, the wider the application range of the distributed optical fiber vibration sensor is, and more kinds of signals can be effectively detected.
The frequency response range in the conventional phi-OTDR technique is mainly affected by the scanning frequency f c of the electro-absorption modulator, both of which satisfy the nyquist sampling theorem. In order to avoid overlapping of backward Rayleigh scattering echoes of each detection light pulse, the next detection light pulse needs to wait for the end of the backward Rayleigh scattering echo of the last detection light pulse before the detection light pulse is injected into the sensing optical fiber, so that the injection frequency of the detection light pulse is limited by the length of the sensing optical fiber, resulting in a frequency response rangeV g is the group velocity of light transmitted in an optical fiber, i.e. as described in the background section, there is the problem that the sensing distance and the frequency response bandwidth are constrained to each other.
As described above, the application has N detection pulse lights with wavelengths in the whole sensing optical fiber, so that the pulse repetition frequency can be increased to be N times of the original pulse repetition frequency under the condition that the sensing optical fiber has the same length, thereby increasing the frequency response bandwidth to be N times of the original pulse repetition frequency, and further leading the distributed optical fiber vibration sensor to have better performance in the aspects of sensing distance and frequency response bandwidth.
Based on the performance indexes of the distributed optical fiber vibration sensor, relevant performance parameters in the distributed optical fiber vibration sensor can be reasonably designed, so that the performance indexes meet the design requirements.
In another embodiment, the distributed optical fiber vibration sensor includes an optical fiber amplifier and M sensing channels (respectively denoted as sensing channels 1 to M), please refer to fig. 2, the structure of each sensing channel is the same and as shown in fig. 1, the sensing optical fibers in each sensing channel are respectively arranged at different spatial positions of the region to be detected for vibration, and the integer parameter M is not less than 2. The input end of the optical fiber amplifier is connected with the output end of the electroabsorption modulator, each output end of the 1*M coupler contained in the optical fiber amplifier is respectively connected with the signal input ends of M transmission channels, and the signal output ends of all the transmission channels are connected with the control processor. The control processor synthesizes vibration detection signals output by the M photoelectric detectors and outputs a vibration induction result based on the phi-OTDR technology.
In the embodiment of the structure shown in fig. 2, the plurality of sensing channels respectively collect backward rayleigh scattering echoes in the plurality of sensing optical fibers, so as to realize the effect of multi-path sensing. Along with the complexity of the application environment and the processing task, the single-path sensing is often limited by the detection capability of the single-path sensing, and only partial data can be obtained at a certain moment, so that the comprehensive target description cannot be obtained. Meanwhile, faults, error accumulation and the like of the single-path sensing structure can bring influence to the system, and the robustness is low. The embodiment is expanded into a multi-path sensing structure, so that the control processor synthesizes vibration detection signals output by the M photoelectric detectors to obtain a final vibration sensing result, the vibration sensing result is easy to realize, and the anti-interference capability, the stability and the reliability and the fault tolerance of the system can be greatly improved on the premise of not increasing the cost of the system.
The information fusion of the multi-path sensing is generally divided into three layers of structure division, namely data layer fusion, feature layer fusion and decision layer fusion, in order to reduce the requirements of the communication bandwidth and data processing of the system, and in consideration of the very high requirement of the distributed optical fiber vibration sensor on real-time performance, the multi-path sensing information fusion is completed by adopting the decision layer fusion mode, and the vibration detection signals output by the control processor and the M photoelectric detectors are synthesized, and the vibration sensing results are output based on the phi-OTDR technology and comprise:
Firstly, independently making identification judgment for each path of sensing, and obtaining an inductor result based on the phi-OTDR technology according to the vibration detection signal output by each photoelectric detector, wherein the inductor result is used for indicating whether the target position in the vibration area to be detected has vibration or not. In actual implementation, the traditional OTDR positioning algorithm adopts a difference method, a disturbance curve waveform is obtained by calculating the differences of backward Rayleigh scattering echoes at different moments, a disturbance threshold is preset, vibration can be determined when the value of the disturbance curve waveform reaches the disturbance threshold, otherwise, vibration is determined not to occur, and a sensor result is obtained.
And then carrying out decision fusion on the sensor results of each path of sensing to obtain a final vibration sensing result. In one embodiment, the decision fusion of sensor results of each path of sensing is realized by using D-S evidence theory, which comprises the following steps:
(1) An identification framework is constructed.
The D-S evidence theory belongs to the field of information fusion, and an identification framework is a complete set consisting of a plurality of mutually incompatible propositions, and represents all possible answers to a certain question. In the application, the constructed recognition frame comprises 4 mutually exclusive propositions respectively recorded as propositions A 1, propositions A 2, propositions A 3 and propositions A 4: proposition a indicates that the target position is actually vibrating and the sensor result indicates vibrating, proposition a 2 indicates that the target position is actually vibrating and the sensor result indicates no vibrating, proposition a 3 indicates that the target position is actually non-vibrating and the sensor result indicates vibrating, proposition a 4 indicates that the target position is actually non-vibrating and the sensor result indicates no vibrating.
(2) The basic probabilities of the individual sensor results corresponding to the individual propositions under the recognition framework are then initialized. As described above, the basis for determining each sensor channel is a threshold method, and a preset disturbance threshold is used to determine whether vibration occurs. The thresholding method takes on the functions of detecting disturbances and screening out disturbances, so if the disturbance threshold is set too low, there are cases of false detection of vibrations. If the disturbance threshold is set too high, it may lead to a condition of missed detection vibrations. The multi-path sensing scheme of the application well solves the problems: the randomness of the noise makes the probability that the same noise peak appears in the same position of the multipath at the same time very small, so when the multipath scheme selects the disturbance threshold, the disturbance threshold can be reduced as much as possible to improve the sensitivity.
In one embodiment, for any integer parameter 1.ltoreq.j.ltoreq.M, the method of initializing the base probability includes the following:
when the j-th sensor result indicates that the target position has vibration, m j(A2)＝mj(A4) =0 is determined, m j(A1) and m j(A3)＝1-mj(A1) are determined according to the real-time signal-to-noise ratio, and the larger the signal-to-noise ratio is, the larger the value of m p(A1) is. One way is according to Real-time signal to noise ratio/>M j(A1 is obtained in the value interval mapped to (0.5, 1).
When the jth sensor result indicates that the target position has no vibration, m j(A1)＝mj(A3) =0 is determined, m j(A4) and m j(A2)＝1-mj(A4) are determined according to the real-time signal-to-noise ratio, and the larger the signal-to-noise ratio is, the larger the value of m j(A4) is, one way is thatReal-time signal to noise ratio/>M j(A4 is obtained in the value interval mapped to (0.8,1). In the multi-path sensing structure of the application, as described above, the disturbance threshold can be set as small as possible to improve the sensitivity, so that the probability of missed detection is low, and the problem A 2 indicates the condition of missed detection, so that the value interval of the problem A 2 is (0,0.2) and the value is small.
For example, please refer to the following table, in which when m=4, the basic probability of initializing each sensor result to correspond to each proposition under the recognition frame is as follows:
The degree of trust assigned to each proposition is called the basic probability assignment BPA, also called the m-function. The trust function Bel (a i) of any proposition a i represents the sum of the basic probabilities of all subsets of propositions a i. The likelihood function PI (a i) represents the sum of the base probabilities that the intersection with proposition a i is not null. Thus Bel (A i) reflects the degree of trust in proposition A i, PI (A i) is an uncertainty measure that may hold for proposition A i, in fact, [ Bel (A i),PI(Ai) ] represents the uncertainty interval of proposition A i, [0, ael (A i) ] represents the support evidence interval of proposition A i, [0, PI (A i) ] represents the fitting interval of proposition A i, [ PI (A i), 1] represents the rejection evidence interval of proposition A i.
(3) And carrying out information fusion on the M sensor results according to the identification frame based on the D-S evidence theory to obtain a vibration sensing result, wherein the vibration sensing result is used for indicating whether the target position has vibration or not.
After the basic probability distribution function of any two sensor results (two independent evidence sources) is determined, a new basic probability distribution function reflecting fusion information generated by the combined action of the two sensor results can be calculated by utilizing a Dempster combination rule, and then information fusion is realized.
However, the D-S evidence theory also has a certain defect, and when a plurality of sensor results are in high conflict, such as a certain sensing channel fails, the fusion based on the traditional D-S evidence theory easily interferes with the final result. Therefore, in one embodiment, the method includes pre-correcting the data before fusion, calculating similarity information by using consistency of a plurality of sensor results and the measure, further obtaining weights of the sensor results, and finally reasonably distributing combinations and weights of the sensor results in a weighted average mode, wherein the method includes:
according to the similarity between any two inductor results in the M inductor results, the reliability weight of each inductor result is calculated, and the basic probability of each inductor result corresponding to the proposition is weighted and averaged according to the reliability weight of each inductor result, so that the weighted average probability of each proposition is obtained.
Wherein obtaining the weighted average probability of each proposition comprises:
First, the weight of any i-th proposition A i is determined M j(Ai) is the basic probability corresponding to the ith proposition A i in the jth sensor result, the integer parameter 1 is less than or equal to j is less than or equal to M, and the integer parameter 1 is less than or equal to i is less than or equal to 4.
Second, the similarity between any j-th sensor result m j and g-th sensor result m g is calculated by combining the weight of each propositionM g(Ai) is the basic probability corresponding to the ith proposition A i in the g-th sensor result, and the integer parameter 1 is more than or equal to g is more than or equal to M.
Third, calculating the weight of any jth sensor resultAnd carrying out normalization processing to obtain the credibility weight/>, of the j-th sensor result
Fourth, each sensor result is weighted and averaged according to the credibility weight of each sensor result, and the weighted average probability of any ith proposition A i is determined
In the example of the table, the weighted average probabilities of the four propositions calculated using the above procedure are m (a 1)＝0.996、m(A2)＝0、m(A3)＝0.04、m(A4) =0, respectively. It can be seen that, although in this example, the third sensor result collides with the other three sensor results, after the above-mentioned process is modified in advance, the evidence of the collision can be removed, and the other mutually supported information is fused, so that a more uniform and reliable determination can be obtained.
And then, carrying out information fusion on the M sensor results according to the identification frame based on the D-S evidence theory according to the weighted average probability of each proposition to obtain a vibration sensing result. By adding the step of pre-correction, a more accurate vibration sensing result can be obtained.
According to the method, the anti-interference performance of multi-path sensing combined evidence fusion is utilized to make up for the false alarm condition caused by the fact that the disturbance threshold can be reduced, when only an individual sensing channel senses vibration, the weight of a sensing sub-result given to the sensing channel in the fusion process is very low, so that the final fusion result is not affected too much, and the influence of a small part of sensing channels caused by noise can be screened out through an algorithm.
After the D-S evidence theory is used to obtain the vibration sensing result to determine whether the target position vibrates, in a further embodiment, when the vibration sensing result is determined to be used to indicate that the target position vibrates, the vibration detection signals output by the photodetectors are respectively input into the vibration pattern recognition model to obtain the vibration patterns of the vibration detection signals output by the vibration pattern recognition model, and the voting mechanism is used to vote on the vibration patterns of the vibration detection signals to determine the final vibration pattern, where the obtained vibration sensing result is also used to indicate the final vibration pattern. The vibration mode identification model is obtained in advance based on end-to-end convolutional neural network training. The embodiment can realize two-stage vibration detection, the first-stage vibration detection utilizes the D-S evidence theory to carry out multi-path sensing information fusion to judge whether the target position has vibration, and the second-stage vibration detection is more accurate only when the target position has vibration, so that a specific vibration mode is identified.
The vibration pattern recognition model sequentially comprises the following components from input to output: the stacked feature extraction layers, the expansion layer, the full connection layer and the classifier, each feature extraction layer sequentially comprises a convolution layer, an activation function and a pooling layer, please refer to the network architecture of fig. 3:
The convolution layer is used for learning local features in time, a series of feature sequences are generated by carrying out convolution operation on input data through a group of learnable one-dimensional convolution check, and meanwhile, one convolution layer can carry out convolution operation on the input data through a plurality of convolution check, so that feature extraction at different angles is realized, and finally, a multi-channel feature map is generated. In performing the convolution process, the number of convolution filters may be set, with different initial vectors for different filters to extract the various desired features.
The fitting ability of the linear function is limited, so a nonlinear layer, i.e. an activation function, is added at each feature extraction layer. In one embodiment, reLU is employed with an activation function. When the optimizer calculates the gradient dip, the ReLU avoids the problem of gradient extinction. Meanwhile, sparsity in a convolution network is guaranteed, and compared with other activation functions, training time is remarkably shortened.
The pooling layer is an operation of reducing the size of the feature map, by which the size of the feature map can be reduced while important information is maintained. Such dimension reduction helps to reduce the number of parameters and calculations in the model, improve the computational efficiency and reduce the risk of overfitting. While the pooling layer is somewhat robust to slight input variations. In one embodiment, the pooling layer employs a maximum pooling operation.
The unfolding layer is used for unfolding the data output by the feature extraction layer into a one-dimensional structure so as to connect the full-connection layers. And finally outputting a final result through the full-connection layer, converting the output of the full-connection layer into category probability through a classifier, and selecting the category with the highest probability as the vibration mode obtained through recognition. In one embodiment, the classifier employs softmax.
The above is only a preferred embodiment of the present application, and the present application is not limited to the above examples. It is to be understood that other modifications and variations which may be directly derived or contemplated by those skilled in the art without departing from the spirit and concepts of the present application are deemed to be included within the scope of the present application.
Claims (7)
1. A distributed optical fiber vibration sensor based on a time division multiplexing technology, which is characterized by comprising a tunable laser, an electric absorption modulator, a sensing channel and a control processor;
the control processor is connected with and controls the tunable laser, and the tunable laser outputs continuous laser light containing N wavelengths;
the electroabsorption modulator is used for modulating continuous laser generated by the tunable laser into detection pulse light with N wavelengths and inputting the detection pulse light into the signal input end of the sensing channel;
The sensing channel comprises sensing optical fibers, N wavelength division demultiplexers, N circulators and photoelectric detectors, wherein the sensing optical fibers are arranged in a vibration area to be detected, and the sensing optical fibers comprise N continuous optical fiber sections, and N is more than or equal to 2; in the sensing channel, the input end of a first wavelength division multiplexer is connected with the signal input end of the sensing channel, for any integer parameter 1N-1, the first output end of an nth wavelength division multiplexer is connected with the input end of an (n+1) th wavelength division multiplexer through a delay optical fiber with the same length as that of an nth optical fiber section, and the first output end of the nth wavelength division multiplexer is suspended; the second output end of the nth wavelength division demultiplexer is connected with the first port of the nth circulator, the second port of the nth circulator is connected with the optical fiber incident end of the nth optical fiber section, and the optical fiber emergent end of the nth optical fiber section is connected with the third port of the (n+1) th circulator; the optical fiber emergent end of the Nth optical fiber section is suspended, the third port of the first circulator is connected with the photoelectric detector, and the signal output end of the photoelectric detector is connected with the signal output end of the sensing channel; each wavelength division multiplexer corresponds to one wavelength, each wavelength division multiplexer outputs detection pulse light with the corresponding wavelength through a second output end, outputs detection pulse light with other wavelengths through a first output end, the second output ends of the N wavelength division multiplexers respectively output N detection pulse light with different wavelengths, and signals in each circulator are sequentially transmitted according to a first port, a second port and a third port;
The distributed optical fiber vibration sensor comprises an optical fiber amplifier and M sensing channels, sensing optical fibers in each sensing channel are respectively distributed at different spatial positions of a region to be detected for vibration, and integer parameter M is more than or equal to 2; the input end of the optical fiber amplifier is connected with the output end of the electroabsorption modulator, each output end of a 1*M coupler contained in the optical fiber amplifier is respectively connected with signal input ends of M transmission channels, the signal output ends of the transmission channels are all connected with the control processor, the control processor synthesizes vibration detection signals output by the M photoelectric detectors and outputs a vibration induction result based on the phi-OTDR technology, and the vibration induction result is used for indicating vibration disturbance suffered by a sensing optical fiber in the sensing channel and comprises the following components: obtaining an inductor result based on a phi-OTDR technology according to the vibration detection signal output by each photoelectric detector, wherein the inductor result is used for indicating whether the target position in the region to be detected to vibrate is vibrated or not; building an identification framework, wherein the identification framework comprises 4 mutually exclusive propositions respectively: the target position is actually vibrated and the sensor result indicates that there is vibration, the target position is actually vibrated and the sensor result indicates that there is no vibration, the target position is actually not vibrated and the sensor result indicates that there is no vibration; initializing basic probabilities of each sensor result corresponding to each proposition under the identification frame, and carrying out information fusion on M sensor results according to the identification frame based on a D-S evidence theory to obtain vibration sensing results, wherein the vibration sensing results are used for indicating whether the target position has vibration or not;
The information fusion of the M sensor results to obtain a vibration sensing result further comprises: when the vibration sensing result is determined to be used for indicating that the target position vibrates, respectively inputting vibration detection signals output by all photoelectric detectors into a vibration mode identification model to obtain vibration modes of all the vibration detection signals output by the vibration mode identification model, voting the vibration modes of all the vibration detection signals by adopting a voting mechanism to determine a final vibration mode, and then the obtained vibration sensing result is also used for indicating the final vibration mode; the vibration mode identification model is obtained in advance based on end-to-end convolutional neural network training.
2. The distributed optical fiber vibration sensor according to claim 1, wherein the information fusion of the M sensor results based on the D-S evidence theory according to the identification frame to obtain a vibration sensing result comprises:
According to the similarity between any two inductor results in the M inductor results, calculating to obtain the credibility weight of each inductor result, and carrying out weighted average on the basic probability of each inductor result corresponding to the proposition according to the credibility weight of each inductor result to obtain the weighted average probability of each proposition;
And carrying out information fusion on the M sensor results according to the identification frame based on the D-S evidence theory according to the weighted average probability of each proposition to obtain a vibration sensing result.
3. The distributed fiber optic vibration sensor according to claim 2 wherein the deriving a weighted average probability of each proposition comprises:
determining any ith proposition Weights/>，/>Is the j-th sensor result corresponding to the i-th proposition/>The integer parameter 1 is more than or equal to j is more than or equal to M, and the integer parameter 1 is more than or equal to i is more than or equal to 4;
calculating any j-th sensor result by combining weights of all propositions And g-th sensor result/>Similarity of (2)，/>Is the g-th sensor result corresponding to the i-th proposition/>The integer parameter is 1-g-M;
Calculating the weight of any jth sensor result Normalizing to obtain the credibility weight/>, of the j-th sensor result；
Weighting and averaging each sensor result according to the credibility weight of each sensor result to determine any ith propositionWeighted average probability/>。
4. The distributed fiber optic vibration sensor according to claim 1 wherein initializing the base probabilities that each sensor result corresponds to each proposition under the recognition framework comprises, for any j-th sensor result:
determining when the j-th sensor result indicates that the target position has vibration Determining/> based on real-time signal-to-noise ratioAnd/>The greater the signal-to-noise ratio,/>The larger the value is;
determining when the jth sensor result indicates that the target position is not vibrating Determining/> based on real-time signal-to-noise ratioAnd/>The greater the signal-to-noise ratio,/>The larger the value is;
wherein, proposition Indicating that the target position is actually vibrated and the sensor result indicates that the target position is vibrated, proposition/>Indicating that the target position is actually vibrated and the sensor result indicates no vibration, proposition/>Indicating that the target position is practically free from vibration and the sensor result indicates that the target position has vibration, proposition/>The target position is actually free of vibration, the sensor result indicates no vibration, and the integer parameter 1 is more than or equal to j and less than or equal to M.
5. A distributed optical fiber vibration sensor according to claim 4 wherein,
When the jth sensor result indicates that the target position has vibration, the method is as followsReal-time signal to noise ratio/>Mapping to/>Obtained in the value interval of (1)/>；
When the jth sensor result indicates that the target position is not vibrating, the method is as followsReal-time signal to noise ratio/>Mapping to/>Obtained in the value interval of (2)。
6. The distributed fiber optic vibration sensor according to claim 1, wherein the vibration pattern recognition model comprises, in order from input to output: the feature extraction system comprises a plurality of feature extraction layers, an unfolding layer, a full-connection layer and a classifier which are stacked, wherein each feature extraction layer sequentially comprises a convolution layer, an activation function and a pooling layer.
7. The distributed fiber optic vibration sensor according to claim 6, wherein the activation function employs a ReLU and the classifier employs a softmax.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311132994.8A CN117109720B (en) | 2023-09-04 | 2023-09-04 | Distributed optical fiber vibration sensor based on time division multiplexing technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311132994.8A CN117109720B (en) | 2023-09-04 | 2023-09-04 | Distributed optical fiber vibration sensor based on time division multiplexing technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117109720A CN117109720A (en) | 2023-11-24 |
| CN117109720B true CN117109720B (en) | 2024-04-26 |
Family
ID=88799902
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311132994.8A Active CN117109720B (en) | 2023-09-04 | 2023-09-04 | Distributed optical fiber vibration sensor based on time division multiplexing technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117109720B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101650216A (en) * | 2009-09-10 | 2010-02-17 | 上海华魏光纤传感技术有限公司 | Improved carpet type monitoring system and monitoring method thereof |
| CN108760302A (en) * | 2018-05-08 | 2018-11-06 | 南京风电科技有限公司 | A kind of on-line monitoring and fault diagnosis system of wind power generating set bearing |
| CN109556703A (en) * | 2018-11-27 | 2019-04-02 | 电子科技大学 | A kind of distributed sound wave detection system based on time-division multiplex technology |
| CN111811637A (en) * | 2020-08-12 | 2020-10-23 | 桂林电子科技大学 | Vehicle vibration identification device with multi-path information fusion and judgment method |
| CN112580425A (en) * | 2020-10-13 | 2021-03-30 | 浙江工业大学 | Water pump fault classification method based on neural network and D-S evidence theory |
| CN112762970A (en) * | 2021-03-09 | 2021-05-07 | 冉曾令 | High-performance distributed optical fiber sensing system and method |
| CN114676413A (en) * | 2022-02-22 | 2022-06-28 | 江苏大学 | Wearable device identity verification method through reprogrammable vibration signal |
| CN217083950U (en) * | 2022-03-28 | 2022-07-29 | 中国计量大学 | Full-space distributed measuring device based on DAS system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102018124435A1 (en) * | 2018-10-03 | 2020-04-09 | Nkt Photonics Gmbh | Distributed measuring device |
| CN110160573B (en) * | 2019-07-08 | 2022-03-25 | 山东省科学院激光研究所 | Escholtz ultrafast modulation pulse scanning laser and distributed optical fiber sensing system |
-
2023
- 2023-09-04 CN CN202311132994.8A patent/CN117109720B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101650216A (en) * | 2009-09-10 | 2010-02-17 | 上海华魏光纤传感技术有限公司 | Improved carpet type monitoring system and monitoring method thereof |
| CN108760302A (en) * | 2018-05-08 | 2018-11-06 | 南京风电科技有限公司 | A kind of on-line monitoring and fault diagnosis system of wind power generating set bearing |
| CN109556703A (en) * | 2018-11-27 | 2019-04-02 | 电子科技大学 | A kind of distributed sound wave detection system based on time-division multiplex technology |
| CN111811637A (en) * | 2020-08-12 | 2020-10-23 | 桂林电子科技大学 | Vehicle vibration identification device with multi-path information fusion and judgment method |
| CN112580425A (en) * | 2020-10-13 | 2021-03-30 | 浙江工业大学 | Water pump fault classification method based on neural network and D-S evidence theory |
| CN112762970A (en) * | 2021-03-09 | 2021-05-07 | 冉曾令 | High-performance distributed optical fiber sensing system and method |
| CN114676413A (en) * | 2022-02-22 | 2022-06-28 | 江苏大学 | Wearable device identity verification method through reprogrammable vibration signal |
| CN217083950U (en) * | 2022-03-28 | 2022-07-29 | 中国计量大学 | Full-space distributed measuring device based on DAS system |
Non-Patent Citations (4)
| Title |
|---|
| φ-OTDR Signal Identification Method Based on Multimodal Fusion;Zhang HZ, et al;《SENSORS》;20221130;第22卷(第22期);第1-12页 * |
| Φ-OTDR型分布式全光纤传感器研究进展;施羿;封皓;曾周末;;自动化仪表(07);第70-74、79页 * |
| 多传感器信息融合在Φ―OTDR***中的应用;李灏;杨德伟;李立京;钟翔;孙昌君;;传感器与微***;35(6);第139-141页 * |
| 改进的D-S证据理论在Φ-OTDR三路监测***中的应用;张文霞;邵洪峰;王夏霄;于佳;;半导体光电(01);第105-108、112页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117109720A (en) | 2023-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8923663B2 (en) | Distributed fibre optic sensing | |
| US6909094B2 (en) | System and method for terahertz imaging using a single terahertz detector | |
| AU2010252746B2 (en) | Optical sensor and method of use | |
| CN113049084B (en) | Attention mechanism-based Resnet distributed optical fiber sensing signal identification method | |
| CN105784101B (en) | Based on optical fiber weak reflection spot battle array Fizeau interference distributed vibration sensing systems and method | |
| JP2013506828A (en) | Phase-based detection | |
| US5557400A (en) | Multiplexed sensing using optical coherence reflectrometry | |
| CN107515253B (en) | The non-contact quantitative Diagnosis method of girder construction defect based on convolutional neural networks | |
| US20230186001A1 (en) | Method and system for simulating an event | |
| WO2023069333A1 (en) | Few-mode rayleigh-based distributed fiber sensor for simultaneous temperature and strain sensing | |
| Lalam et al. | Recent development in artificial neural network based distributed fiber optic sensors | |
| CN117109720B (en) | Distributed optical fiber vibration sensor based on time division multiplexing technology | |
| Abdelli et al. | Branch identification in passive optical networks using machine learning | |
| US20230086130A1 (en) | Optic signal receiver, optic communication system and optic communication method | |
| WO2016105401A1 (en) | Apparatus and method for characterization of fbg reflector array | |
| CN112504433B (en) | Temperature self-reference sensitivity tunable optical fiber distributed vibration detection device and method | |
| Lyu et al. | A light-weight neural network for marine acoustic signal recognition suitable for fiber-optic hydrophones | |
| Hayle et al. | Self-healing integration of fiber/FSO communication and sensor network for improving survivability | |
| Abdelli et al. | Faulty branch identification in passive optical networks using machine learning | |
| CN112484837B (en) | Optical fiber space positioning system and implementation method thereof | |
| WO2022029412A1 (en) | Distributed fibre optic sensing | |
| CN111750793B (en) | Intensity demodulation strain sensing method based on neural network | |
| KR20200069772A (en) | Optical cable observing system and optical cable observing method | |
| WO2023056079A1 (en) | Outdoor application of distributed fiber optic sensing/acoustic sensing | |
| CN116279649A (en) | Track safety detection system and method based on improved AlexNet |
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 |