CN103235953B - A kind of method of optical fiber distributed perturbation sensor pattern recognition - Google Patents
A kind of method of optical fiber distributed perturbation sensor pattern recognition Download PDFInfo
- Publication number
- CN103235953B CN103235953B CN201310125329.6A CN201310125329A CN103235953B CN 103235953 B CN103235953 B CN 103235953B CN 201310125329 A CN201310125329 A CN 201310125329A CN 103235953 B CN103235953 B CN 103235953B
- Authority
- CN
- China
- Prior art keywords
- frequency
- time period
- output signal
- distribution characteristic
- pattern recognition
- 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.)
- Expired - Fee Related
Links
Landscapes
- Alarm Systems (AREA)
- Emergency Alarm Devices (AREA)
Abstract
A kind of method that the invention discloses optical fiber distributed perturbation sensor pattern recognition, the method includes: according to time period output signal I to optical fiber distributed perturbation sensor1T () carries out M sampling, it is thus achieved that the N number of sampled point divided according to the time period, wherein, and N M;Zero-crossing rate F (i) of sampled point in the calculating i-th time period, and calculate equivalent frequency f (i) in current slot according to this zero-crossing rate F (i), it is thus achieved that the frequency of current slot this sensor output signal interior and Annual distribution characteristic;Frequency and Annual distribution characteristic according to described sensor output signal carry out pattern recognition.By using method disclosed by the invention to improve recognition efficiency height and accuracy, reduce false alarm rate.
Description
Technical field
The present invention relates to optical fiber distributed perturbation sensor technical field, particularly relate to a kind of optical fiber distributed perturbation sensor mould
Formula knows method for distinguishing.
Background technology
The disturbance of any point on sensor fibre can be detected, identify and position by optical fiber distributed perturbation sensor,
By its monitoring distance length, can position, without important technical advantage such as outfield power, can be widely applied to circumference peace
Anti-, oil-gas pipeline early warning, communication line monitoring and large scale structure monitoring.
The workflow of optical fiber distributed perturbation sensor is detection external disturbance signal, and disturbing signal is carried out pattern knowledge
Not, it is determined whether make alarm decision.The service behaviour of optical fiber distributed perturbation sensor can be entered by three probability levels
Line description: the probability of optical fiber distributed perturbation sensor detection external disturbance is detectivity, and alert event is not made report
The alert probability judged is false dismissed rate, and the probability that non-alert event carries out alarm decision is false alarm rate.
The key issue of limit fibre distributed perturbation sensor further genralrlization application at present is still can not to realize disturbing to external world
Dynamic pattern recognition fast and accurately, causes false alarm rate in actual application higher.
Inventor finds that in carrying out innovation and creation prior art is primarily present following defect:
1) algorithm for pattern recognition based on analysis of spectrum.Signal is mainly carried out in Fu by algorithm for pattern recognition based on analysis of spectrum
Leaf transformation (or wavelet transformation), differentiates signal characteristic at frequency domain (or wavelet field).This algorithm deposits amount of calculation
Relatively big, response time is long, simultaneously need to demarcate sensor under various circumstances, practicality is the highest.
2) dynamic time warping (DTW).This algorithm thought based on dynamic programming (DP), is used for solving to pass
The sense matching problem that signal is different in size or signal is the most corresponding with template, is to occur relatively morning, more classics in speech recognition
A kind of algorithm.But, in order to obtain preferable pattern recognition rate, DTW algorithm needs a huge template base, tool
There is operand big, the shortcoming that arithmetic speed is slow;And it is mainly used in voice signal identification, it is not suitable for optical fiber distributed type and disturbs
The pattern recognition of dynamic sensor.
3) recessive Markov model (HMM) algorithm for pattern recognition.Hidden markov models (HMM) is Ma Er
Can the one of husband's chain, its state can not observe directly, and can only be arrived by observation vector sequence inspection.HMM is a kind of
Models based on probability statistics, need substantial amounts of data to carry out model learning, and in real process, multi-signal collection can
Difficulty can be there will be, cause model to set up error, thus have influence on signal identification;And it is mainly used in voice signal knowledge
Not, the pattern recognition of optical fiber distributed perturbation sensor it is not suitable for.
4) mode identification method based on artificial neural network (ANN).The mode identification method of ANN is with mathematical model
Imictron is movable, is a kind of information processing system set up based on imitating cerebral nerve network structure and function.But
It is, mode identification method based on ANN, the most sensitive to signal time information, even if the most corresponding time delay is same
Signal, the most also can produce Different Results, cause recognition failures.
5) algorithm based on classification.This algorithm is typically to regard input signal as one multi-C vector, real by training
Now hyperspace is divided, thus reach the purpose of classification.Modal sorting technique is SVM, but SVM side
Method and ANN identify have identical shortcoming, are all can not to extend to the time well to adjudicate.
Existing mode identification method is not to be specifically designed for optical fiber distributed perturbation sensor be designed and optimize, to biography
The effect that sense signal carries out pattern recognition is unsatisfactory.Therefore, pattern high in the urgent need to a kind of efficiency, that accuracy is high is known
Other algorithm carries out pattern recognition to Fibre Optical Sensor signal.
Summary of the invention
A kind of method that it is an object of the invention to provide optical fiber distributed perturbation sensor pattern recognition, improves recognition efficiency
Height and accuracy, reduce false alarm rate.
It is an object of the invention to be achieved through the following technical solutions:
A kind of method of optical fiber distributed perturbation sensor pattern recognition, the method includes:
According to time period output signal I to optical fiber distributed perturbation sensor1T () carries out M sampling, it is thus achieved that according to the time period
The N number of sampled point divided, wherein, N M;
Zero-crossing rate F (i) of sampled point in the calculating i-th time period, and calculate in current slot according to described zero-crossing rate F (i)
Equivalent frequency f (i), it is thus achieved that the frequency of current slot this sensor output signal interior and Annual distribution characteristic;
Frequency and Annual distribution characteristic according to described sensor output signal carry out pattern recognition.
As seen from the above technical solution provided by the invention, by sensor output signal is carried out time division, and
The zero-crossing rate utilizing each time period characterizes the equivalent frequency in this time period, thus obtains the frequency of sensor output signal
Rate and Annual distribution characteristic;Using frequency and Annual distribution characteristic as the key foundation of pattern recognition, make mode decision,
Have the advantages that computing is simple, efficiency is high, accuracy is high and false alarm rate is low.
Accompanying drawing explanation
In order to be illustrated more clearly that the technical scheme of the embodiment of the present invention, required use in embodiment being described below
Accompanying drawing is briefly described, it should be apparent that, the accompanying drawing in describing below is only some embodiments of the present invention, for
From the point of view of those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to obtain according to these accompanying drawings
Other accompanying drawings.
The flow chart of the method for a kind of optical fiber distributed perturbation sensor pattern recognition that Fig. 1 provides for the embodiment of the present invention;
Disturbance event frequency that Fig. 2 provides for the embodiment of the present invention and the schematic diagram of Annual distribution characteristic;
Pretreated disturbance event frequency that Fig. 3 provides for the embodiment of the present invention and the schematic diagram of Annual distribution characteristic.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clearly and completely
Describe, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments.Based on
Embodiments of the invention, it is every other that those of ordinary skill in the art are obtained under not making creative work premise
Embodiment, broadly falls into protection scope of the present invention.
The frequency spectrum of signal has reacted signal average energy distribution in each frequency component within a period of time.By signal by time domain
It is transformed into frequency domain to be analyzed, some only the most unobtainable parameter of time domain can be obtained, signal analysis is had important meaning
Justice.But, only signal is carried out frequency-domain analysis, the time-domain information of signal can be lost, such as the variation tendency of signal.By upper
State two kinds of frequencies combined and average or equivalent frequency that Annual distribution characteristic can reflect in the signal short time becomes in time
The characteristic changed.
The amplitude versus time distribution difference of different disturbing signals, the strong and weak change difference of i.e. different disturbances, therefore, monitoring sensing
The frequency of device signal and Annual distribution characteristic can effectively identify disturbance character.
Embodiment
The flow chart of the method for a kind of optical fiber distributed perturbation sensor pattern recognition that Fig. 1 provides for the embodiment of the present invention.As
Shown in Fig. 1, the method mainly comprises the steps:
Step 11, according to time period output signal I to optical fiber distributed perturbation sensor1T () carries out M sampling, it is thus achieved that root
The N number of sampled point divided according to the time period.
The embodiment of the present invention carries out pattern recognition based on frequency and Annual distribution characteristic.Accordingly, it would be desirable to optical fiber distributed type is disturbed
Output signal I of dynamic sensor1T () was sampled according to the time period.Such as, according to the time period carry out M(more than 0 the most whole
Number) secondary sampling, it is thus achieved that the N(divided according to the time period is more than or equal to the positive integer of M) individual sampled point;Under normal circumstances,
The number of sampling is the most identical every time, and the most each time period sampled point number is N/M.
Step 12, calculate zero-crossing rate F (i) of sampled point in the i-th time period, and calculate current time according to this zero-crossing rate F (i)
Equivalent frequency f (i) in Duan, it is thus achieved that the frequency of current slot this sensor output signal interior and Annual distribution characteristic.
The present embodiment utilizes zero-crossing rate F (i) to characterize frequency and the Annual distribution characteristic of the equivalent frequency in this time period, i.e. signal
Can be extracted by segmentation zero-crossing rate and obtain.
Zero-crossing rate F (i) represents sampled point sampled point number in number of times Z (i) and this time period of zero crossing in the i-th time period
Ratio, is considered as sinusoidal signal by sensor output signal, then equivalent frequency f (i) within each time period is:
Wherein, fsFor sample frequency.
After this simplified formula, can obtain:The i.e. zero-crossing rate of sensor output signal each time period
F (i) can characterize equivalent frequency f (i) in this time period, such that it is able to it is special to obtain zero-crossing rate (equivalent frequency)-Annual distribution
Property.
Step 13, frequency and Annual distribution characteristic according to described sensor output signal carry out pattern recognition.
The embodiment of the present invention can be by disturbance thing in the frequency of described sensor output signal and Annual distribution characteristic and data base
The frequency of part compares with Annual distribution characteristic, select frequency and Annual distribution characteristic closest to disturbance event as knowledge
Other result.
As in figure 2 it is shown, the frequency of different disturbance events and Annual distribution characteristic exist significant difference, wherein, A represents percussion
Optical cable;When the time period is 10, its frequency reaches peak value (0.25MHZ);Optical cable is rocked in the expression of B;When the time period it is
During 6-10, its frequency reaches minimum (0MHZ);C represents and pulls optical cable;When the time period is 90, its frequency exceedes
0.35MHZ;D represents vibrations near optical cable;Its frequency maintains relatively low state (less than 0.05MHZ).
In real work, for improving recognition efficiency and accuracy, disturbance event frequency and time in data base can be divided
Cloth characteristic carries out pretreatment;For example, with reference to Fig. 3, by setting frequency and the Annual distribution characteristic of threshold control signal.So
After, the signal frequency obtained according to step 11-step 12 is processed by identical threshold value with Annual distribution characteristic, then
Compared with disturbance event frequency pretreated with data base and Annual distribution characteristic, determine immediate a kind of disturbance thing
Part is recognition result.
The embodiment of the present invention by carrying out time division by sensor output signal, and utilizes the zero-crossing rate of each time period
Characterize the equivalent frequency in this time period, thus obtain frequency and the Annual distribution characteristic of sensor output signal;By frequency
With Annual distribution characteristic as the key foundation of pattern recognition, make mode decision, have that computing is simple, efficiency is high, accurate
The feature that exactness is high and false alarm rate is low.
Through the above description of the embodiments, those skilled in the art it can be understood that to above-described embodiment permissible
Realized by software, it is also possible to the mode adding necessary general hardware platform by software realizes.Based on such reason
Solving, the technical scheme of above-described embodiment can embody with the form of software product, and this software product can be stored in one
In individual non-volatile memory medium (can be CD-ROM, USB flash disk, portable hard drive etc.), including some instructions with so that
One computer equipment (can be personal computer, server, or the network equipment etc.) performs the present invention, and each is implemented
Method described in example.
The above, the only present invention preferably detailed description of the invention, but protection scope of the present invention is not limited thereto,
Any those familiar with the art in the technical scope of present disclosure, the change that can readily occur in or replace
Change, all should contain within protection scope of the present invention.Therefore, protection scope of the present invention should be with claims
Protection domain is as the criterion.
Claims (2)
1. the method for an optical fiber distributed perturbation sensor pattern recognition, it is characterised in that the method includes:
According to time period output signal I to optical fiber distributed perturbation sensor1T () carries out M sampling, it is thus achieved that according to the time period
The N number of sampled point divided, wherein, N M;
Calculate zero-crossing rate F (i) of sampled point in the i-th time period, and according to this zero-crossing rate F (i) calculate in current slot etc.
Effect frequency f (i), it is thus achieved that the frequency of current slot this sensor output signal interior and Annual distribution characteristic;
Frequency and Annual distribution characteristic according to described sensor output signal carry out pattern recognition;
Wherein, in the described calculating i-th time period, zero-crossing rate F (i) of sampled point includes: calculate sampled point warp in the i-th time period
Number of times Z (i) of zero crossing and the ratio of sampled point number in this time period;
The described formula calculating equivalent frequency f (i) in current slot according to described zero-crossing rate F (i) includes:
Wherein, fsFor sample frequency.
Method the most according to claim 1, it is characterised in that the described frequency according to described sensor output signal
Carry out pattern recognition with Annual distribution characteristic to include:
Frequency and the Annual distribution characteristic of described sensor output signal were divided with frequency and the time of disturbance event in data base
Cloth characteristic compares, select frequency and Annual distribution characteristic closest to disturbance event as recognition result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310125329.6A CN103235953B (en) | 2013-04-11 | 2013-04-11 | A kind of method of optical fiber distributed perturbation sensor pattern recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310125329.6A CN103235953B (en) | 2013-04-11 | 2013-04-11 | A kind of method of optical fiber distributed perturbation sensor pattern recognition |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103235953A CN103235953A (en) | 2013-08-07 |
CN103235953B true CN103235953B (en) | 2016-12-28 |
Family
ID=48883992
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310125329.6A Expired - Fee Related CN103235953B (en) | 2013-04-11 | 2013-04-11 | A kind of method of optical fiber distributed perturbation sensor pattern recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103235953B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103903270B (en) * | 2014-04-09 | 2017-02-08 | 东南大学 | Regularized valid characteristic section selecting method of optical fiber link monitoring signals |
CN104035396B (en) * | 2014-04-18 | 2016-08-17 | 重庆大学 | Distributed Activity recognition method based on wireless sensor network |
CN103994784A (en) * | 2014-05-26 | 2014-08-20 | 天津大学 | Distributed optical fiber sensing positioning method based on zero crossing point analysis |
CN104964699B (en) * | 2015-05-22 | 2017-09-08 | 北京交通大学 | Disturbance determination methods and device based on φ OTDR optical fiber distributed perturbation sensors |
CN105222885B (en) * | 2015-06-19 | 2018-07-24 | 北方工业大学 | Optical fiber vibration detection method and device |
CN110135283A (en) * | 2019-04-25 | 2019-08-16 | 上海大学 | The signal recognition method of optical fiber perimeter defence system based on FastDTW algorithm |
CN111208142B (en) * | 2019-08-01 | 2022-03-22 | 北京航空航天大学 | Crack damage quantitative detection method based on dynamic time warping correlation characteristics |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102045120A (en) * | 2010-10-29 | 2011-05-04 | 成都九洲电子信息系统有限责任公司 | Vibration signal identification method for optical fiber perimeter system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7999930B2 (en) * | 2008-10-07 | 2011-08-16 | Corning Incorporated | Detection system and optical fiber for use in such system |
-
2013
- 2013-04-11 CN CN201310125329.6A patent/CN103235953B/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102045120A (en) * | 2010-10-29 | 2011-05-04 | 成都九洲电子信息系统有限责任公司 | Vibration signal identification method for optical fiber perimeter system |
Non-Patent Citations (1)
Title |
---|
数字信号处理在连续分布式光纤传感定位系统中的应用;柴岩凤;《中国优秀硕士学位论文全文数据库 信息科技辑》;20121015(第10期);第5、17-33页 * |
Also Published As
Publication number | Publication date |
---|---|
CN103235953A (en) | 2013-08-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103235953B (en) | A kind of method of optical fiber distributed perturbation sensor pattern recognition | |
CN104240455B (en) | A kind of disturbance event recognition methods in distribution type fiber-optic pipeline safety early warning system | |
CN103617684B (en) | Interference-type optical fiber circumference vibrating intruding recognizer | |
CN109034046B (en) | Method for automatically identifying foreign matters in electric energy meter based on acoustic detection | |
Li et al. | Damage detection of bridges using response of vehicle considering road surface roughness | |
CN109489800A (en) | A kind of disturbance event recognition methods in distribution optic cable vibration safety pre-warning system | |
CN104729667A (en) | Method for recognizing disturbance type in a distributed optical fiber vibration sensing system | |
Vercauteren et al. | A clustering method to characterize intermittent bursts of turbulence and interaction with submesomotions in the stable boundary layer | |
CN104463359A (en) | Dredging operation yield prediction model analysis method based on BP neural network | |
CN111398679B (en) | Sub-synchronous oscillation identification and alarm method based on PMU (phasor measurement Unit) | |
CN105606198A (en) | Fiber vibration sensing system two-order signal feature extraction and determining method | |
Yan et al. | Bearing fault diagnosis via a parameter-optimized feature mode decomposition | |
CN108229553A (en) | A kind of OTDR curve datas analysis method | |
CN109002810A (en) | Model evaluation method, Radar Signal Recognition method and corresponding intrument | |
CN103207931A (en) | Sub-synchronous oscillation mode identification method based on milli molar (MM) and autoregressive moving average (ARMA) algorithm | |
CN104964736B (en) | Optical fiber invasion vibration source identification method based on time-frequency characteristic maximum expected classification | |
Ma et al. | An intelligent classification framework for complex PQDs using optimized KS-transform and multiple fusion CNN | |
CN117606577A (en) | Method and system for monitoring switching flow of reclaimed water and tap water in real time | |
US10585128B2 (en) | Noise spectrum analysis for electronic device | |
CN103400021A (en) | KDSNE hydroelectric generating set noise source detection method based on heavy tailed distribution | |
Liu et al. | Application of Hilbert-Huang Transform and SVM to Coal Gangue Interface Detection. | |
CN104730384A (en) | Power disturbance identification and localization method based on incomplete S transformation | |
CN115577249A (en) | Transformer acoustic signal identification method, system and medium with multi-view feature fusion | |
CN112329535B (en) | CNN-based quick identification method for low-frequency oscillation modal characteristics of power system | |
CN111524036B (en) | Transient power quality disturbance classification method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20161228 Termination date: 20170411 |
|
CF01 | Termination of patent right due to non-payment of annual fee |