CN109489800A - A kind of disturbance event recognition methods in distribution optic cable vibration safety pre-warning system - Google Patents
A kind of disturbance event recognition methods in distribution optic cable vibration safety pre-warning system Download PDFInfo
- Publication number
- CN109489800A CN109489800A CN201811538869.6A CN201811538869A CN109489800A CN 109489800 A CN109489800 A CN 109489800A CN 201811538869 A CN201811538869 A CN 201811538869A CN 109489800 A CN109489800 A CN 109489800A
- Authority
- CN
- China
- Prior art keywords
- disturbance event
- disturbance
- optic cable
- index
- original signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses the disturbance event recognition methods in a kind of distributed optic cable vibration safety pre-warning system for belonging to technical field of optical fiber sensing to carry out multiple cumulative mean when there are disturbance event to optical path sampled signal and wavelet de-noising is handled；Characteristic value is extracted to the sampled signal after noise reduction process；20 characteristic values of extraction are sent into gradient boosted tree and carry out modeling of class, Tree Classifier is promoted using gradient and obtains disturbance event type；In the new disturbance event type of appearance or in the case where gradient promotes the disturbance event type appearance mistake that Tree Classifier obtains, by being modified to the disturbance event type stored in model database, it realizes the lower interaction incremental learning of artificial supervision, Tree Classifier progress on-line training is promoted to gradient according to revised disturbance event type.The present invention enough can accurately obtain disturbance event type.
Description
Technical field
The present invention relates to technical field of optical fiber sensing, disturbing in specifically a kind of distribution optic cable vibration safety pre-warning system
Dynamic event recognition method.
Background technique
With the development of society, the progress of optical communication technique, communications optical cable has become the most important basis of modern society
One of facility, much hundreds of of the backbone communication interruption of optical cables as caused by third party's breakage in installation, cause to be difficult to retrieve every year
Economic loss, the protection for communications optical cable is gradually taken seriously.However in the geographical environment of communications optical cable approach major part
It all in urban road two sides, is rolled, the factors such as various engineering activities are influenced, is formed by communication O&M cable laying, oversize vehicle
Multiple vibration source.The real-time monitoring for having had been provided with overlength currently based on the distributed optical cable early warning system that coherent rayleigh scatters is pre-
How the positioning accuracy of alert ability and superelevation identifies that multiple vibration source becomes distributed optical cable safety under complicated earth surface environment
The urgent need of early warning system.Its difficult point is: 1, with the intelligent recognition ability to the event schema under complicated earth surface environment；
2, with for reply external environment variation and caused by new type event incremental learning ability.
Distributed communication optical cable safety pre-warning system is based on coherent rayleigh scattering principle, and sensing principle is as shown in Figure 1, logical
The phase change for crossing scattered light signal caused by measurement disturbs along communications optical cable positions disturbance location.Pulsed light exists
It is propagated in optical fiber and generates scattering, if acting on sensing fiber ring without disturbance event, the pulsed light and scattering light exported does not have
There is phase difference, phenomenon will not be interfered.If there is a disturbance event to act on certain point on sensor fibre, the phase of light is scattered
It changes at this point and generates interference phenomenon, the interference signal that photodetector receives scattering light can reflect the phase of scattering light
Potential difference, and obtain the precise position information and disturbance information of disturbance point.
Existing similar distributed optical fiber vibration sensing system, is all based on deterministic models, i.e. vibration signal model.Chang Cai
With features such as position, energy, frequency spectrum, LC, kurtosis, the string hair vibration detection of certain a priori known and identification are achieved certain
Effect.But for the long range early warning system under complicated earth surface environment, prior information can not often be obtained, and be vibrated almost always
Occur parallel.It is difficult accurately to detect vibration source and identification vibration source using single features model.It can be only applied under specified environment,
Classify to specified disturbance event and distinguish, incremental learning can not be carried out again with the variation of response environment and new type disturbance event
Occur, lacks versatility.Disturbance event recognition methods in existing distributed optical fiber vibration sensing system also has using support
The classifier of vector machine becomes more dependent on the order of accuarcy of training sample data using the classifier of support vector machines, when some sample
This error is larger, can produce a very large impact to whole training effect, further influence classification accuracy.
Summary of the invention
The purpose of the present invention is to provide the disturbance event identification sides in a kind of distributed optic cable vibration safety pre-warning system
Method can accurately obtain disturbance in the relatively more similar situation of the destructive insidents characteristics of signals such as environmental noise interference and mechanical execution
Event type, to solve the problems mentioned in the above background technology.
To achieve the above object, the invention provides the following technical scheme:
A kind of disturbance event recognition methods in distribution optic cable vibration safety pre-warning system, comprising the following steps:
S1 is fitted original signal baseline using least square method, by making the difference to original signal and the baseline curve of fitting
Mode, remove signal in DC component and trend term component, extract effective oscillating component, to removal trend term after original letter
Number digital filtering and difference pretreatment are carried out, when the amplitude of difference is more than given threshold, there are disturbance events for judgement, then to original
Beginning signal carries out wavelet de-noising processing；
S2, to wavelet de-noising treated all the way sampled signal extract characteristic value, characteristic value includes peak value, peak-to-peak value, mistake
Zero rate, pulse index, waveform index, kurtosis, peak index, barycenter of frequency spectrum, gross energy, zero-crossing rate, crosses direct current at root amplitude
Rate, signal length, signal energy, kurtosis and low frequency energy, medium-high frequency energy, frequency-domain waveform index, frequency domain spectra entropy, frequency domain peak value
Index；
20 characteristic values of extraction are sent into gradient and promote Tree Classifier by S3, are obtained and are disturbed using gradient promotion Tree Classifier
Dynamic event category.
As a further solution of the present invention: peak index:
Waveform index:
X is original signal, XiFor i-th of data of original signal, XmaxOriginal signal numerical value maximum value, XminFor original letter
Number value minimum value,For original signal mean value.
As a further solution of the present invention: frequency domain peak index:
Frequency domain spectra entropy:
Frequency-domain waveform index:
Y is frequency spectrum of the original signal after Fourier changes, Y in formulamaxFor frequency spectrum maximum value, yiFor original signal
I-th of spectral magnitude.
As a further solution of the present invention: gradient promotes Tree Classifier and training sample sky is arranged according to the number of characteristic value
Between size, the size of every class disturbance event training sample is at least 20 times of characteristic value number.
As a further solution of the present invention: intrusion event in sensing optic cable being positioned according to Rayleigh scattering principle
Method is the position L according to following formula calculation perturbation point apart from optical cable starting point:
L=Δ t*c/2n
C is that light velocity Δ t is to issue pulse to the Rayleigh curve changed time is received in vacuum, and n is that fiber core is rolled over
Penetrate rate.
As a further solution of the present invention: classifier uses GBDT algorithm.
As a further solution of the present invention: if disturbance event classification is vehicle pass-through, not alarming, by the feature of extraction
Value is directly stored in property data base storage；It alarms if disturbance event classification is artificial construction or mechanical execution event, it will
The characteristic value of extraction is stored in property data base.
The disturbance event classification appearance mistake that Tree Classifier obtains is promoted in the new disturbance event classification of appearance or in gradient
In the case where accidentally, it is modified by field verification and to the disturbance event label stored in database, Lai Shixian human-computer interaction
Incremental learning promotes Tree Classifier to gradient according to modified disturbance event classification and carries out on-line training.
Compared with prior art, the beneficial effects of the present invention are:
1, the present invention combines Time-Frequency Analysis and statistical analysis means, utilizes the peak value of noise reduction process post-sampling signal, peak
Peak value, zero-crossing rate, root amplitude, pulse index, waveform index, kurtosis, peak index, barycenter of frequency spectrum, gross energy, zero-crossing rate,
Cross direct current rate, signal length, signal energy, kurtosis and low frequency energy, medium-high frequency energy, frequency-domain waveform index, frequency domain spectra entropy, frequency
Domain peak index can be carried out effectively for oversize vehicle right of way signal and mechanical execution disturbing signal characteristic as feature vector
Classification, improves the accuracy of mechanical execution pattern-recognition；
2, compared with other existing classification methods, the present invention promotes the design of decision tree classifier, energy by gradient
It is enough effectively to analyze characteristic component more complicated, to need the combination of multiple characteristic values that differentiate disturbance event progress mode
Identification.Gradient promotion decision tree classifier classification speed is fast, and calculation amount is small, is easily converted to classifying rules；Classification accuracy is high,
The regular accuracy excavated from decision tree classifier is high and is easy to understand；
3, the design of decision tree classifier is promoted by gradient, so that sorter model has the incremental learning of human-computer interaction
Function not only can correct existing tagsort system in the use process of system, be continuously improved and know to existing perturbation mode
Other accuracy, and the variation of environment or the appearance of new disturbance event are coped with, improve the versatility of system.
Detailed description of the invention
Fig. 1 is that the present invention is based on coherent rayleigh scatter-type distribution vibrating sensing schematic diagrams.
Fig. 2 is present invention disturbance positioning and algorithm for pattern recognition flow chart.
Fig. 3 is decision tree classifier algorithm effect figure of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Embodiment 1
As shown in Figure 1, distribution type fiber-optic pipeline safety early warning system of the invention, including super-narrow line width light source, acousto-optic tune
Device processed, circulator, sensor fibre, photodetector, high-speed collection card, computer.
The signal of photodetector output is sent into high-speed collection card and carries out sampling acquisition sampled signal and calculate signal spy
Sign, carries out disturbance event pattern-recognition using computer, carries out disturbance event recognition methods in a computer as shown in Fig. 2, tool
Body includes the following steps:
S1, it is fitted original signal baseline using least square method, by making the difference to original signal and the baseline curve of fitting
Mode, remove signal in DC component and trend term component, extract effective oscillating component.To the original letter after removal trend term
Number digital filtering and difference pretreatment are carried out, when the amplitude of difference is more than given threshold, there are disturbance events for judgement, then to original
Beginning signal carries out wavelet de-noising processing.
S2, to wavelet de-noising treated all the way sampled signal extract characteristic value, characteristic value includes peak value, peak-to-peak value, mistake
Zero rate, pulse index, waveform index, kurtosis, peak index, barycenter of frequency spectrum, gross energy, zero-crossing rate, crosses direct current at root amplitude
Rate, signal length, signal energy, kurtosis and low frequency energy, medium-high frequency energy, frequency-domain waveform index, frequency domain spectra entropy, frequency domain peak value
Index.
Peak index:
Waveform index:
X is original signal, XiFor i-th of data of original signal, XmaxOriginal signal numerical value maximum value, XminFor original letter
Number value minimum value,For original signal mean value.
Frequency domain peak index:
Frequency domain spectra entropy:
Frequency-domain waveform index:
Y is frequency spectrum of the original signal after Fourier changes, Y in formulamaxFor frequency spectrum maximum value, yiFor original signal
I-th of spectral magnitude.
S3, the characteristic value for obtaining step S2 are sent into gradient and promote decision tree classifier, obtain disturbance event classification.
Gradient promotes decision tree classifier and uses top-down recursive fashion, is divided according to characteristic value disturbance event
Class carries out the comparison of characteristic value in the internal node of decision tree in assorting process, and is judged inside this according to different characteristic value
The downward branch of node, until reaching some leaf node, to find classification belonging to the disturbance event.In each of decision tree
Portion's node represents the primary test to a certain characteristic value, and each edge represents a test process, and each leaf node represents some class
Other distribution.
If disturbance event classification is environmental background noise, system is not alarmed, and is only sent directly into the characteristic value of signal
Database purchase.
If disturbance event classification is intrusion event, alarm, according to Rayleigh scattering principle to entering in sensing optic cable
The method that the event of invading is positioned is the position L according to following formula calculation perturbation point apart from optical cable starting point:
L=Δ t*c/2n
C is that light velocity Δ t is to issue pulse to the Rayleigh curve changed time is received in vacuum, and n is that fiber core is rolled over
Penetrate rate.
If disturbance event classification is not one of known intrusion event classification, to the same of intrusion event locating alarming
When, feature vector by original signal and after treatment is sent directly into database, in domain expert or duty personnel to it
After having been stored new category disturbance event label in the database, then the design of decision tree classifier is re-started, realized
Human-computer interaction incremental learning.
It mainly include to the incremental learning of old classification new samples and to category in the incremental learning link of classifier of the present invention
In two processes of incremental learning of the new samples of new category.Since the acquisition of sample is a constantly accumulation, abundant and perfect mistake
Journey.Newly-increased sample and old sample are re-started instruction when there is new samples addition by higher nicety of grading in order to obtain altogether
Practice.For decision tree classifier, since its construction algorithm is to carry out on the basis of grasping global information, and work as newly-increased
After adding one kind, the relationship between every class may all be changed, and old tree construction can for new global information
Decision tree classifier can be redesigned in the case in a manner of incremental learning to be optimal solution and non-optimal solution.
Training sample space size is arranged according to the number of characteristic value in decision tree classifier, to guarantee classifier design effect
Fruit, the size of general every class disturbance event training sample are at least 20 times of the number of characteristic value.
This system classifies to live disturbance event using known initial characteristics database, the basis in implementation process
The environmental information of field calibration feedback, and the warning message verified.By live duty personnel or domain expert to initial characteristics number
It is re-flagged and is modified according to library.Step up alarm accuracy rate.
Embodiment 2
In instances, data acquisition is evergreen to Xinzhou section communications optical cable, test optical fibre cables long 45 from Hubei Wuhan commmunication company
Kilometer, part of communications optical cable are laid in underground piping, and optic cable is aerial.This system sample frequency is 100M, sampling
Number N is 6400*2048.
Disturbance event is divided into touching cable, vehicle pass-through interference, mechanical execution three types, system operation according to use demand
In the process, discovery ambient noise mostlys come from travels with the parallel highway oversize vehicle of optical cable, since oversize vehicle is travelled to edge
Linear light cable generates disturbance triggering alarm.Staff combines oversize vehicle to travel the information such as disturbing signal waveform and disturbance location,
The nearly trimestral disturbing signal data for picking up from test optical fibre cables are analyzed, and disturbance event is carried out in system database
Category label.Since system classifies to disturbing signal using 20 characteristic values, to guarantee classifier design effect, every kind is disturbed
Dynamic event category number of training is set as 2000, carries out decision tree classifier training, has demarcated classification training sample characteristic value
It is as shown in table 1:
Table 1 has demarcated classification training sample characteristic value
Feature 1, feature 2 to feature 20 are followed successively by vibration segment peak value, peak-to-peak value, zero-crossing rate, root amplitude, pulse refer to
Mark, waveform index, kurtosis, peak index, barycenter of frequency spectrum, gross energy, zero-crossing rate, cross direct current rate, signal length, signal energy,
Kurtosis and low frequency energy, medium-high frequency energy, frequency domain spectra entropy, frequency-domain waveform index, frequency domain peak index.
It can get decision tree classifier as shown in Figure 3 using classification training sample has been demarcated, as shown in figure 3, participating in dividing
Category feature selection is characterized 1, feature 2, feature 4, feature 5, feature 8, feature 10, feature 17, feature 18, feature 20, right respectively
Induction signal peak value, peak-to-peak value, root amplitude, pulse index, kurtosis, barycenter of frequency spectrum, frequency domain spectra entropy, frequency-domain waveform index and frequency
Domain peak index.
For the validity for verifying system disturbance pattern recognition function, the signal of three classes disturbance trigger event is selected respectively, and
It calculates characteristic of division value to be sent into decision tree classifier shown in Fig. 3, the feature of each node of decision tree classifier to signal
Value is determined, the branch downward from the internal node is judged according to different characteristic value, until reaching some leaf node, to look for
To classification belonging to the disturbance event, gradient promoted decision tree classifier output result be respectively " touching cable ", " vehicle pass-through ",
" mechanical execution ", it was demonstrated that the validity of system disturbance event schema recognition methods.Wherein, the classification of three classes disturbance trigger event
Characteristic value is as shown in table 2:
2 three classes of table disturb trigger event characteristic of division value
The gradient finally finished using training is promoted the disturbance that decision tree classifier picks up from test optical fibre cables to recent months and believed
Number classification verifying is carried out, and carry out live real-time event identification test, mechanical execution pattern-recognition rate is in final system operation
90% or more, reach desired design effect, meets actual application demand.To also demonstrate distribution proposed by the invention
The reliability and validity of event recognition method in fiber-optic vibration early warning system.
Unspecified part of the present invention belongs to common sense well known to those skilled in the art.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case where without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims
Variation is included within the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.
In addition, it should be understood that although this specification is described in terms of embodiments, but not each embodiment is only wrapped
Containing an independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art should
It considers the specification as a whole, the technical solutions in the various embodiments may also be suitably combined, forms those skilled in the art
The other embodiments being understood that.
Claims (7)
1. the disturbance event recognition methods in a kind of distribution optic cable vibration safety pre-warning system, which is characterized in that including following
Step:
S1 is fitted original signal baseline, the side made the difference by the baseline curve to original signal and fitting using least square method
Formula, remove signal in DC component and trend term component, extract effective oscillating component, to removal trend term after original signal into
Row digital filtering and difference pretreatment, when the amplitude of difference is more than given threshold, there are disturbance events for judgement, then to original letter
Number carry out wavelet de-noising processing；
S2, to wavelet de-noising treated all the way sampled signal extract characteristic value, characteristic value include peak value, peak-to-peak value, zero-crossing rate,
Root amplitude, waveform index, kurtosis, peak index, barycenter of frequency spectrum, gross energy, zero-crossing rate, crosses direct current rate, signal at pulse index
Length, signal energy, kurtosis and low frequency energy, medium-high frequency energy, frequency-domain waveform index, frequency domain spectra entropy, frequency domain peak index；
20 characteristic values of extraction are sent into gradient and promote Tree Classifier by S3, are promoted Tree Classifier using gradient and are obtained disturbance thing
Part classification.
2. the disturbance event recognition methods in a kind of distributed optic cable vibration safety pre-warning system according to claim 1,
It is characterized in that,
Peak index:
Waveform index:
X is original signal, XiFor i-th of data of original signal, XmaxOriginal signal numerical value maximum value, XminFor original signal number
It is worth minimum value,For original signal mean value.
3. the disturbance event recognition methods in a kind of distributed optic cable vibration safety pre-warning system according to claim 1,
It is characterized in that,
Frequency domain peak index:
Frequency domain spectra entropy:
Frequency-domain waveform index:
Y is frequency spectrum of the original signal after Fourier changes, Y in formulamaxFor frequency spectrum maximum value, yiIt is i-th of original signal
Spectral magnitude.
4. the disturbance event recognition methods in a kind of distributed optic cable vibration safety pre-warning system according to claim 1,
It is characterized in that, gradient, which promotes Tree Classifier, is arranged training sample space size, every class disturbance event according to the number of characteristic value
The size of training sample is at least 20 times of characteristic value number.
5. the disturbance event recognition methods in a kind of distributed optic cable vibration safety pre-warning system according to claim 1,
It is characterized in that, being according to following formula meter according to the method that Rayleigh scattering principle positions intrusion event in sensing optic cable
Calculate position L of the disturbance point apart from optical cable starting point:
L=△ t*c/2n
C is that light velocity △ t is to issue pulse to the Rayleigh curve changed time is received in vacuum, and n is fiber core refractive index.
6. the disturbance event recognition methods in a kind of distributed optic cable vibration safety pre-warning system according to claim 1,
It is characterized in that, classifier uses GBDT algorithm.
7. the disturbance event identification side in -6 any distributed optic cable vibration safety pre-warning systems according to claim 1
Method, which is characterized in that if disturbance event classification is vehicle pass-through, does not alarm, the characteristic value of extraction is directly stored in characteristic
It is stored according to library；It alarms if disturbance event classification is artificial construction or mechanical execution event, the characteristic value of extraction is stored in
Property data base.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811538869.6A CN109489800A (en) | 2018-12-14 | 2018-12-14 | A kind of disturbance event recognition methods in distribution optic cable vibration safety pre-warning system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811538869.6A CN109489800A (en) | 2018-12-14 | 2018-12-14 | A kind of disturbance event recognition methods in distribution optic cable vibration safety pre-warning system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109489800A true CN109489800A (en) | 2019-03-19 |
Family
ID=65710413
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811538869.6A Pending CN109489800A (en) | 2018-12-14 | 2018-12-14 | A kind of disturbance event recognition methods in distribution optic cable vibration safety pre-warning system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109489800A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110440901A (en) * | 2019-08-13 | 2019-11-12 | 郑州信大先进技术研究院 | A kind of distributed optical fiber vibration sensing localization method and device based on pulse accumulation |
| CN110458071A (en) * | 2019-08-01 | 2019-11-15 | 北京邮电大学 | A kind of fiber-optic vibration signal characteristic abstraction and classification method based on DWT-DFPA-GBDT |
| CN110598312A (en) * | 2019-09-09 | 2019-12-20 | 武汉安保通科技有限公司 | Underground vibration event type identification method and system |
| CN110928237A (en) * | 2019-12-20 | 2020-03-27 | 华中科技大学 | Vibration signal-based numerical control machining center flutter online identification method |
| CN111160106A (en) * | 2019-12-03 | 2020-05-15 | 上海微波技术研究所（中国电子科技集团公司第五十研究所） | Method and system for extracting and classifying optical fiber vibration signal features based on GPU |
| CN111397850A (en) * | 2020-04-20 | 2020-07-10 | 深圳市特发信息股份有限公司 | Vibration waveform and database sample analysis system for optical cable safety |
| CN112245728A (en) * | 2020-06-03 | 2021-01-22 | 北京化工大学 | Respirator false positive alarm signal identification method and system based on integrated tree |
| CN112542046A (en) * | 2020-12-07 | 2021-03-23 | 无锡科晟光子科技有限公司 | Early warning monitoring method for long-distance pipeline heavy vehicle based on DAS |
| CN112883521A (en) * | 2021-01-12 | 2021-06-01 | 中国科学院声学研究所南海研究站 | Seabed photoelectric composite cable external force invasion monitoring system applied to seabed observation network |
| CN112985574A (en) * | 2021-02-26 | 2021-06-18 | 电子科技大学 | High-precision classification identification method for optical fiber distributed acoustic sensing signals based on model fusion |
| CN113063489A (en) * | 2021-02-24 | 2021-07-02 | 戴天智能科技（上海）有限公司 | Vibration detection system with AI self-learning algorithm |
| CN113722481A (en) * | 2021-08-23 | 2021-11-30 | 国家计算机网络与信息安全管理中心 | Text multi-event detection method and device based on category and instance enhancement |
| CN113781727A (en) * | 2021-08-25 | 2021-12-10 | 广州东部发展燃气有限公司 | Intelligent gas sign pile |
| CN114004263A (en) * | 2021-12-29 | 2022-02-01 | 四川大学 | Large-scale equipment working condition diagnosis and prediction method based on feature fusion conversion |
| CN114061638A (en) * | 2021-10-26 | 2022-02-18 | 广东工业大学 | Brillouin optical time domain reflectometer and phase demodulation method and device thereof |
| CN116186642A (en) * | 2023-04-27 | 2023-05-30 | 山东汇英信息科技有限公司 | Distributed optical fiber sensing event early warning method based on multidimensional feature fusion |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102564476A (en) * | 2011-10-28 | 2012-07-11 | 北京航空航天大学 | Multipoint disturbance positioning method |
| CN102997945A (en) * | 2011-09-16 | 2013-03-27 | 北京航空航天大学 | Multipoint disturbance positioning method of fiber-distributed disturbance sensor |
| CN103090961A (en) * | 2013-02-01 | 2013-05-08 | 华中科技大学 | Disturbance source positioning method of distributed type optical fiber sensing system |
| US8610886B2 (en) * | 2009-09-28 | 2013-12-17 | At&T Intellectual Property I, L.P. | Long distance optical fiber sensing system and method |
| CN103954450A (en) * | 2014-05-19 | 2014-07-30 | 重庆交通大学 | Bearing life degradation performance evaluation index construction method based on main component analysis |
| CN104215323A (en) * | 2014-08-11 | 2014-12-17 | 中国人民解放军空军工程大学 | Method for determining sensitivity of each sensor in mechanical equipment vibrating sensor network |
| CN104240455A (en) * | 2014-08-07 | 2014-12-24 | 北京航天控制仪器研究所 | Method for identifying disturbance event in distributed type optical fiber pipeline security early-warning system |
| CN104729667A (en) * | 2015-03-25 | 2015-06-24 | 北京航天控制仪器研究所 | Method for recognizing disturbance type in a distributed optical fiber vibration sensing system |
| CN105184319A (en) * | 2015-09-02 | 2015-12-23 | 同方威视技术股份有限公司 | Fiber perimeter intrusion signal identification method and apparatus, and perimeter intrusion alarm system |
| CN105632059A (en) * | 2015-12-29 | 2016-06-01 | 天津大学 | Distributed optical fiber perimeter security system |
| CN106204993A (en) * | 2016-08-15 | 2016-12-07 | 成都市和平科技有限责任公司 | A kind of resident's circumference intrusion alarm system and alarm method thereof |
| CN107369282A (en) * | 2017-08-23 | 2017-11-21 | 武汉理工光科股份有限公司 | A kind of buried COTDR circumferences intrusion alarm system and signal processing method |
| CN108154065A (en) * | 2016-12-02 | 2018-06-12 | 光子瑞利科技（北京）有限公司 | A kind of circumference early warning fiber-optic vibration signal acquisition and the method for denoising |
-
2018
- 2018-12-14 CN CN201811538869.6A patent/CN109489800A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8610886B2 (en) * | 2009-09-28 | 2013-12-17 | At&T Intellectual Property I, L.P. | Long distance optical fiber sensing system and method |
| CN102997945A (en) * | 2011-09-16 | 2013-03-27 | 北京航空航天大学 | Multipoint disturbance positioning method of fiber-distributed disturbance sensor |
| CN102564476A (en) * | 2011-10-28 | 2012-07-11 | 北京航空航天大学 | Multipoint disturbance positioning method |
| CN103090961A (en) * | 2013-02-01 | 2013-05-08 | 华中科技大学 | Disturbance source positioning method of distributed type optical fiber sensing system |
| CN103954450A (en) * | 2014-05-19 | 2014-07-30 | 重庆交通大学 | Bearing life degradation performance evaluation index construction method based on main component analysis |
| CN104240455A (en) * | 2014-08-07 | 2014-12-24 | 北京航天控制仪器研究所 | Method for identifying disturbance event in distributed type optical fiber pipeline security early-warning system |
| CN104215323A (en) * | 2014-08-11 | 2014-12-17 | 中国人民解放军空军工程大学 | Method for determining sensitivity of each sensor in mechanical equipment vibrating sensor network |
| CN104729667A (en) * | 2015-03-25 | 2015-06-24 | 北京航天控制仪器研究所 | Method for recognizing disturbance type in a distributed optical fiber vibration sensing system |
| CN105184319A (en) * | 2015-09-02 | 2015-12-23 | 同方威视技术股份有限公司 | Fiber perimeter intrusion signal identification method and apparatus, and perimeter intrusion alarm system |
| CN105632059A (en) * | 2015-12-29 | 2016-06-01 | 天津大学 | Distributed optical fiber perimeter security system |
| CN106204993A (en) * | 2016-08-15 | 2016-12-07 | 成都市和平科技有限责任公司 | A kind of resident's circumference intrusion alarm system and alarm method thereof |
| CN108154065A (en) * | 2016-12-02 | 2018-06-12 | 光子瑞利科技（北京）有限公司 | A kind of circumference early warning fiber-optic vibration signal acquisition and the method for denoising |
| CN107369282A (en) * | 2017-08-23 | 2017-11-21 | 武汉理工光科股份有限公司 | A kind of buried COTDR circumferences intrusion alarm system and signal processing method |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458071A (en) * | 2019-08-01 | 2019-11-15 | 北京邮电大学 | A kind of fiber-optic vibration signal characteristic abstraction and classification method based on DWT-DFPA-GBDT |
| CN110440901B (en) * | 2019-08-13 | 2021-08-17 | 郑州信大先进技术研究院 | Distributed optical fiber vibration sensing positioning method and device based on pulse accumulation |
| CN110440901A (en) * | 2019-08-13 | 2019-11-12 | 郑州信大先进技术研究院 | A kind of distributed optical fiber vibration sensing localization method and device based on pulse accumulation |
| CN110598312A (en) * | 2019-09-09 | 2019-12-20 | 武汉安保通科技有限公司 | Underground vibration event type identification method and system |
| CN111160106A (en) * | 2019-12-03 | 2020-05-15 | 上海微波技术研究所（中国电子科技集团公司第五十研究所） | Method and system for extracting and classifying optical fiber vibration signal features based on GPU |
| CN111160106B (en) * | 2019-12-03 | 2023-12-12 | 上海微波技术研究所（中国电子科技集团公司第五十研究所） | GPU-based optical fiber vibration signal feature extraction and classification method and system |
| CN110928237A (en) * | 2019-12-20 | 2020-03-27 | 华中科技大学 | Vibration signal-based numerical control machining center flutter online identification method |
| CN111397850A (en) * | 2020-04-20 | 2020-07-10 | 深圳市特发信息股份有限公司 | Vibration waveform and database sample analysis system for optical cable safety |
| CN112245728A (en) * | 2020-06-03 | 2021-01-22 | 北京化工大学 | Respirator false positive alarm signal identification method and system based on integrated tree |
| CN112542046A (en) * | 2020-12-07 | 2021-03-23 | 无锡科晟光子科技有限公司 | Early warning monitoring method for long-distance pipeline heavy vehicle based on DAS |
| CN112542046B (en) * | 2020-12-07 | 2022-03-11 | 无锡科晟光子科技有限公司 | Early warning monitoring method for long-distance pipeline heavy vehicle based on DAS |
| CN112883521A (en) * | 2021-01-12 | 2021-06-01 | 中国科学院声学研究所南海研究站 | Seabed photoelectric composite cable external force invasion monitoring system applied to seabed observation network |
| CN113063489A (en) * | 2021-02-24 | 2021-07-02 | 戴天智能科技（上海）有限公司 | Vibration detection system with AI self-learning algorithm |
| CN113063489B (en) * | 2021-02-24 | 2023-01-13 | 戴天智能科技（上海）有限公司 | Vibration detection system with AI self-learning algorithm |
| CN112985574A (en) * | 2021-02-26 | 2021-06-18 | 电子科技大学 | High-precision classification identification method for optical fiber distributed acoustic sensing signals based on model fusion |
| CN113722481A (en) * | 2021-08-23 | 2021-11-30 | 国家计算机网络与信息安全管理中心 | Text multi-event detection method and device based on category and instance enhancement |
| CN113722481B (en) * | 2021-08-23 | 2023-09-22 | 国家计算机网络与信息安全管理中心 | Text multi-event detection method and device based on category and instance enhancement |
| CN113781727A (en) * | 2021-08-25 | 2021-12-10 | 广州东部发展燃气有限公司 | Intelligent gas sign pile |
| CN114061638A (en) * | 2021-10-26 | 2022-02-18 | 广东工业大学 | Brillouin optical time domain reflectometer and phase demodulation method and device thereof |
| CN114061638B (en) * | 2021-10-26 | 2024-03-26 | 广东工业大学 | Brillouin optical time domain reflectometer and phase demodulation method and device thereof |
| CN114004263A (en) * | 2021-12-29 | 2022-02-01 | 四川大学 | Large-scale equipment working condition diagnosis and prediction method based on feature fusion conversion |
| US11630034B2 (en) | 2021-12-29 | 2023-04-18 | Sichuan University | Method for diagnosing and predicting operation conditions of large-scale equipment based on feature fusion and conversion |
| CN116186642A (en) * | 2023-04-27 | 2023-05-30 | 山东汇英信息科技有限公司 | Distributed optical fiber sensing event early warning method based on multidimensional feature fusion |
| CN116186642B (en) * | 2023-04-27 | 2023-09-08 | 山东汇英光电科技有限公司 | Distributed optical fiber sensing event early warning method based on multidimensional feature fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109489800A (en) | A kind of disturbance event recognition methods in distribution optic cable vibration safety pre-warning system | |
| CN104240455B (en) | A kind of disturbance event recognition methods in distribution type fiber-optic pipeline safety early warning system | |
| CN104729667B (en) | A kind of disturbance kind identification method in distributed optical fiber vibration sensing system | |
| CN108709633B (en) | Distributed optical fiber vibration sensing intelligent safety monitoring method based on deep learning | |
| CN107256635B (en) | Vehicle identification method based on distributed optical fiber sensing in intelligent traffic | |
| CN103971523B (en) | A kind of mountain road traffic safety dynamic early-warning system | |
| CN103617684B (en) | Interference-type optical fiber circumference vibrating intruding recognizer | |
| CN110738856B (en) | Mobile clustering-based urban traffic jam fine identification method | |
| CN108932480A (en) | The study of distributing optical fiber sensing signal characteristic and classification method based on 1D-CNN | |
| CN104217513B (en) | The method improving phase sensitive optical time domain reflectometer identification intrusion event accuracy rate | |
| CN103116957B (en) | A kind of method of optical fiber perimeter security and protection system shielding climate impact | |
| CN105632059A (en) | Distributed optical fiber perimeter security system | |
| CN103235953B (en) | A kind of method of optical fiber distributed perturbation sensor pattern recognition | |
| CN105023379A (en) | Signal identification method of fiber perimeter early-warning system of airport | |
| Sun et al. | Man-made threat event recognition based on distributed optical fiber vibration sensing and SE-WaveNet | |
| CN108801437A (en) | Distributed optical fiber vibration sensing localization method and device based on disturbing signal feature extraction | |
| CN103994815A (en) | Method for recognizing optical fiber vibration source | |
| Papp et al. | Train detection and tracking in optical time domain reflectometry (OTDR) signals | |
| Li et al. | Quickly build a high-precision classifier for Φ-OTDR sensing system based on transfer learning and support vector machine | |
| CN108986363A (en) | Optical fiber security protection intrusion event recognition methods and device based on ARMA modeling | |
| Zhu et al. | A comprehensive bibliometric analysis of signal processing and pattern recognition based on distributed optical fiber | |
| CN109120336B (en) | False alarm prevention and false alarm prevention method based on phase sensitive optical time domain reflection sensor | |
| CN113447727B (en) | Method and device for positioning lightning above oil and gas pipeline based on optical fiber vibration measurement system | |
| CN104964736B (en) | Optical fiber invasion vibration source identification method based on time-frequency characteristic maximum expected classification | |
| CN104916059B (en) | M Z interferometer types airport circumference early warning system mode identification method based on EEMD |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190319 |
|
| RJ01 | Rejection of invention patent application after publication |