CN103235953A - Pattern recognition method for fiber-optic distributed disturbance sensor - Google Patents
Pattern recognition method for fiber-optic distributed disturbance sensor Download PDFInfo
- Publication number
- CN103235953A CN103235953A CN2013101253296A CN201310125329A CN103235953A CN 103235953 A CN103235953 A CN 103235953A CN 2013101253296 A CN2013101253296 A CN 2013101253296A CN 201310125329 A CN201310125329 A CN 201310125329A CN 103235953 A CN103235953 A CN 103235953A
- Authority
- CN
- China
- Prior art keywords
- frequency
- time
- output signal
- time period
- 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.)
- Granted
Links
Images
Landscapes
- Emergency Alarm Devices (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a pattern recognition method for a fiber-optic distributed disturbance sensor. The method includes: sampling an output signal I1 (t) of the fiber-optic distributed disturbance sensor for M times according to periods of time so as to obtain N sample points divided according to the periods of time, N >/=M; calculating zero-crossing rate F(i) of the sample point in the i-th period of time, and calculating equivalent frequency f(i) in the current period of time according to the zero-crossing rate F(i) to obtain frequency and time distribution characteristics of the sensor output signal in the current period of time; and recognizing patterns according to the frequency and time distribution characteristics of the sensor output signal. By the use of the method, recognition efficiency and accuracy are increased, and false alarm rate is reduced.
Description
Technical field
The present invention relates to the optical fiber distributed perturbation sensor technical field, relate in particular to a kind of method of optical fiber distributed perturbation sensor pattern-recognition.
Background technology
Optical fiber distributed perturbation sensor can on the sensor fibre arbitrarily the disturbance of any survey, identify and locate, rely on its monitoring distance to grow, can locate, need not important technical advantage such as outfield power supply, can be widely used in circumference security protection, oil-gas pipeline early warning, communication line monitoring and large scale structure monitoring.
The workflow of optical fiber distributed perturbation sensor is carried out pattern-recognition for surveying the external disturbance signal to disturbing signal, determines whether to make alarm decision.The serviceability of optical fiber distributed perturbation sensor can be described by three probability levels: the probability that optical fiber distributed perturbation sensor is surveyed external disturbance is detectivity, the probability of alert event not being made alarm decision is false dismissed rate, and the probability that non-alert event is carried out alarm decision is the alert rate of mistake.
The key issue that the present distributed perturbation sensor of limit fibre is further applied is still to realize the pattern-recognition fast and accurately of disturbance to external world, and it is higher to cause missing alert rate in the practical application.
The inventor finds that in carrying out innovation and creation mainly there is following defective in prior art:
1) based on the algorithm for pattern recognition of analysis of spectrum.Algorithm for pattern recognition based on analysis of spectrum mainly is that signal is carried out Fourier transform (or wavelet transformation), at frequency domain (or wavelet field) signal characteristic is differentiated.This algorithm deposits that calculated amount is bigger, and the response time is long, need demarcate sensor under varying environment simultaneously, and practicality is not high.
2) dynamic time programming algorithm (DTW).This algorithm is based on the thought of dynamic programming (DP), is used for solving the matching problem that transducing signal is different in size or signal is not corresponding with template, be occur in the speech recognition early, a kind of algorithm of classics comparatively.But in order to obtain pattern-recognition rate preferably, the DTW algorithm needs a huge template base, and it is big to have an operand, the shortcoming that arithmetic speed is slow; And be mainly used in voice signal identification, be not suitable for the pattern-recognition of optical fiber distributed perturbation sensor.
3) recessive Markov model (HMM) algorithm for pattern recognition.Hidden markov models (HMM) is a kind of of Markov chain, and its state can not observe directly, and can only observe by the observation vector sequence.HMM is a kind of model based on probability statistics, need lot of data to carry out model learning, and in the real process, multiple signals collecting may encounter difficulties, and causes model to set up error, thereby has influence on signal identification; And be mainly used in voice signal identification, be not suitable for the pattern-recognition of optical fiber distributed perturbation sensor.
4) based on the mode identification method of artificial neural network (ANN).The mode identification method of ANN is with the activity of mathematical model imictron, is based on imitation cerebral nerve network structure and function and a kind of information handling system of setting up.But, responsive inadequately to signal time information based on the mode identification method of ANN, even have only the same signal of corresponding time-delay, also can produce Different Results sometimes, cause recognition failures.
5) based on classification algorithms.This algorithm is normally regarded input signal as a multi-C vector, realizes hyperspace is divided by training, thereby reaches the purpose of classification.Modal sorting technique is SVM, and identification has identical shortcoming yet the SVM method is with ANN, all is can not extend to the time well to adjudicate.
Existing mode identification method is not to design at optical fiber distributed perturbation sensor specially and optimize, and the effect of transducing signal being carried out pattern-recognition is unsatisfactory.Therefore, press for a kind of efficient height, the high algorithm for pattern recognition of accuracy comes the Fibre Optical Sensor signal is carried out pattern-recognition.
Summary of the invention
The method that the purpose of this invention is to provide a kind of optical fiber distributed perturbation sensor pattern-recognition has improved recognition efficiency height and accuracy, has reduced the alert rate of mistake.
The objective of the invention is to be achieved through the following technical solutions:
A kind of method of optical fiber distributed perturbation sensor pattern-recognition, this method comprises:
According to the output signal I of time period to optical fiber distributed perturbation sensor
1(t) carry out M sampling, obtain N sampled point according to the time period division, wherein, N ≧ M;
Calculate the zero-crossing rate F (i) of sampled point in i time period, and calculate the interior equivalent frequency f (i) of current slot according to described zero-crossing rate F (i), obtain frequency and the time distribution character of this sensor output signal in the current slot;
Frequency and time distribution character according to described sensor output signal carry out pattern-recognition.
As seen from the above technical solution provided by the invention, divide by sensor output signal being carried out the time, and the zero-crossing rate that utilizes each time period characterizes the equivalent frequency in this time period, thereby obtains frequency and the time distribution character of sensor output signal; With frequency and the time distribution character key foundation as pattern-recognition, the pattern of making judges, has that computing is simple, efficient is high, accuracy is high and the low characteristics of the alert rate of mistake.
Description of drawings
In order to be illustrated more clearly in the technical scheme of the embodiment of the invention, the accompanying drawing of required use is done to introduce simply in will describing embodiment below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite of not paying creative work, can also obtain other accompanying drawings according to these accompanying drawings.
The process flow diagram of the method for a kind of optical fiber distributed perturbation sensor pattern-recognition that Fig. 1 provides for the embodiment of the invention;
The disturbance event frequency that Fig. 2 provides for the embodiment of the invention and the synoptic diagram of time distribution character;
The pretreated disturbance event frequency that Fig. 3 provides for the embodiment of the invention and the synoptic diagram of time distribution character.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on embodiments of the invention, those of ordinary skills belong to protection scope of the present invention not making the every other embodiment that obtains under the creative work prerequisite.
The frequency spectrum of signal has reacted signal average energy distribution on each frequency component in a period of time.Signal is transformed into frequency domain analysis by time domain, can obtains some only very difficult parameter that obtains of time domain, significant to signal analysis.Yet, only signal is carried out frequency-domain analysis, can lose the time-domain information of signal, as the variation tendency of signal.Can the average or equivalent frequency time dependent characteristic of reflected signal in the short time with above-mentioned two kinds of frequencies that combine and time distribution character.
The amplitude of different disturbing signals-time distributes different, i.e. the power of different disturbances changes different, and therefore, the frequency of monitors sensor signal and time distribution character can effectively be identified disturbance character.
Embodiment
The process flow diagram of the method for a kind of optical fiber distributed perturbation sensor pattern-recognition that Fig. 1 provides for the embodiment of the invention.As shown in Figure 1, this method mainly comprises the steps:
The embodiment of the invention is carried out pattern-recognition based on frequency and time distribution character.Therefore, need be to the output signal I of optical fiber distributed perturbation sensor
1(t) sample according to the time period.For example, carry out M(greater than 0 positive integer according to the time period) inferior sampling, obtain the N(that divides according to the time period more than or equal to the positive integer of M) individual sampled point; Generally, the number of each sampling is all identical, and then each time period sampled point number is N/M.
The zero-crossing rate F (i) of step 12, i interior sampled point of time period of calculating, and according to the equivalent frequency f (i) in this zero-crossing rate F (i) calculating current slot, obtain frequency and the time distribution character of interior this sensor output signal of current slot.
Present embodiment utilizes zero-crossing rate F (i) to characterize interior equivalent frequency of this time period, and namely the frequency of signal and time distribution character can extract by the segmentation zero-crossing rate and obtain.
The ratio of i interior sampled point of time period of zero-crossing rate F (i) expression sampled point number in the number of times Z of zero crossing (i) and this time period is considered as sinusoidal signal with sensor output signal, and then the equivalent frequency f (i) in each time period is:
Wherein, f
sBe sample frequency.
Behind this simplified formula, can get:
The zero-crossing rate F (i) that is each time period of sensor output signal can characterize interior equivalent frequency f (i) of this time period, thereby can obtain zero-crossing rate (equivalent frequency)-time distribution character.
The embodiment of the invention can be compared frequency and the time distribution character of described sensor output signal with frequency and the time distribution character of disturbance event in the database, select the most approaching disturbance event of frequency and time distribution character as recognition result.
As shown in Figure 2, there are significant difference in the frequency of different disturbance events and time distribution character, and wherein, A represents to knock optical cable; When the time period was 10, its frequency reached peak value (0.25MHZ); Optical cable is rocked in the expression of B; When the time period was 6-10, its frequency reached minimum (0MHZ); C represents to spur optical cable; When the time period was 90, its frequency surpassed 0.35MHZ; D represents to shake near the optical cable; Its frequency dimension is held in low state (being lower than 0.05MHZ).
In real work, for improving recognition efficiency and accuracy, can carry out pre-service to disturbance event frequency in the database and time distribution character; For example, referring to Fig. 3, by frequency and the time distribution character of setting threshold control signal.Then, to handle by identical threshold value with the time distribution character according to the signal frequency that step 11-step 12 obtains, compare with pretreated disturbance event frequency and time distribution character in the database again, determine that immediate a kind of disturbance event is recognition result.
The embodiment of the invention was divided by the time that sensor output signal is carried out, and utilized the zero-crossing rate of each time period to characterize interior equivalent frequency of this time period, thereby obtained frequency and the time distribution character of sensor output signal; With frequency and the time distribution character key foundation as pattern-recognition, the pattern of making judges, has that computing is simple, efficient is high, accuracy is high and the low characteristics of the alert rate of mistake.
Through the above description of the embodiments, those skilled in the art can be well understood to above-described embodiment and can realize by software, also can realize by the mode that software adds necessary general hardware platform.Based on such understanding, the technical scheme of above-described embodiment can embody with the form of software product, it (can be CD-ROM that this software product can be stored in a non-volatile memory medium, USB flash disk, portable hard drive etc.) in, comprise some instructions with so that computer equipment (can be personal computer, server, the perhaps network equipment etc.) carry out the described method of each embodiment of the present invention.
The above; only for the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto, and anyly is familiar with those skilled in the art in the technical scope that the present invention discloses; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.
Claims (4)
1. the method for an optical fiber distributed perturbation sensor pattern-recognition is characterized in that, this method comprises:
According to the output signal I of time period to optical fiber distributed perturbation sensor
1(t) carry out M sampling, obtain N sampled point according to the time period division, wherein, N ≧ M;
Calculate the zero-crossing rate F (i) of sampled point in i time period, and calculate the interior equivalent frequency f (i) of current slot according to this zero-crossing rate F (i), obtain frequency and the time distribution character of this sensor output signal in the current slot;
Frequency and time distribution character according to described sensor output signal carry out pattern-recognition.
2. method according to claim 1 is characterized in that, the zero-crossing rate F (i) of i interior sampled point of time period of described calculating comprises:
Calculate the ratio of sampled point sampled point number in the number of times Z of zero crossing (i) and this time period in i time period.
3. method according to claim 1 and 2 is characterized in that, described formula according to the equivalent frequency f (i) in described zero-crossing rate F (i) the calculating current slot comprises:
Wherein, f
sBe sample frequency.
4. method according to claim 1 is characterized in that, described frequency and time distribution character according to described sensor output signal carries out pattern-recognition and comprise:
Frequency and the time distribution character of described sensor output signal are compared with frequency and the time distribution character of disturbance event in the database, select the most approaching disturbance event of frequency and time distribution character 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 true CN103235953A (en) | 2013-08-07 |
CN103235953B 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) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103903270A (en) * | 2014-04-09 | 2014-07-02 | 东南大学 | Regularized valid characteristic section selecting method of optical fiber link monitoring signals |
CN103994784A (en) * | 2014-05-26 | 2014-08-20 | 天津大学 | Distributed optical fiber sensing positioning method based on zero crossing point analysis |
CN104035396A (en) * | 2014-04-18 | 2014-09-10 | 重庆大学 | Distributed behavior identification method based on wireless sensor network |
CN104964699A (en) * | 2015-05-22 | 2015-10-07 | 北京交通大学 | Disturbance determining method and apparatus based on phi-OTDR fiber distributed type disturbance sensor |
CN105222885A (en) * | 2015-06-19 | 2016-01-06 | 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 |
CN111208142A (en) * | 2019-08-01 | 2020-05-29 | 北京航空航天大学 | Crack damage quantitative detection method based on dynamic time warping correlation characteristics |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100085558A1 (en) * | 2008-10-07 | 2010-04-08 | Xin Chen | Detection System and Optical Fiber for Use in Such System |
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 |
-
2013
- 2013-04-11 CN CN201310125329.6A patent/CN103235953B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100085558A1 (en) * | 2008-10-07 | 2010-04-08 | Xin Chen | Detection System and Optical Fiber for Use in Such System |
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 |
---|
柴岩凤: "数字信号处理在连续分布式光纤传感定位系统中的应用", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103903270A (en) * | 2014-04-09 | 2014-07-02 | 东南大学 | Regularized valid characteristic section selecting method of optical fiber link monitoring signals |
CN103903270B (en) * | 2014-04-09 | 2017-02-08 | 东南大学 | Regularized valid characteristic section selecting method of optical fiber link monitoring signals |
CN104035396A (en) * | 2014-04-18 | 2014-09-10 | 重庆大学 | Distributed behavior identification method based on wireless sensor network |
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 |
CN104964699A (en) * | 2015-05-22 | 2015-10-07 | 北京交通大学 | Disturbance determining method and apparatus based on phi-OTDR fiber distributed type disturbance sensor |
CN105222885A (en) * | 2015-06-19 | 2016-01-06 | 北方工业大学 | Optical fiber vibration detection method and device |
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 |
CN111208142A (en) * | 2019-08-01 | 2020-05-29 | 北京航空航天大学 | Crack damage quantitative detection method based on dynamic time warping correlation characteristics |
Also Published As
Publication number | Publication date |
---|---|
CN103235953B (en) | 2016-12-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103235953A (en) | Pattern recognition method for fiber-optic distributed disturbance sensor | |
Kong et al. | Weakly labelled audioset tagging with attention neural networks | |
Wu et al. | A dynamic time sequence recognition and knowledge mining method based on the hidden Markov models (HMMs) for pipeline safety monitoring with Φ-OTDR | |
Christou et al. | Field measurements of rogue water waves | |
Cui et al. | Wind power ramp event forecasting using a stochastic scenario generation method | |
CN103617684B (en) | Interference-type optical fiber circumference vibrating intruding recognizer | |
Lopac et al. | Detection of non-stationary GW signals in high noise from Cohen’s class of time–frequency representations using deep learning | |
CN104317681A (en) | Behavioral abnormality automatic detection method and behavioral abnormality automatic detection system aiming at computer system | |
Kang et al. | Detecting and classifying events in noisy time series | |
CN104331583B (en) | A kind of Multifractal Modeling method based on Observed sea clutter | |
CN109359693A (en) | A kind of Power Quality Disturbance Classification Method | |
CN102982347B (en) | Method for electric energy quality disturbance classification based on KL distance | |
CN105510970A (en) | Method for obtaining seismic facies optimal classification number | |
CN104966161A (en) | Electric energy quality recording data calculating analysis method based on Gaussian mixture model | |
Ye et al. | A novel integrated method for short-term wind power forecasting based on fluctuation clustering and history matching | |
Dai et al. | Complexity–entropy causality plane based on power spectral entropy for complex time series | |
CN102592593A (en) | Emotional-characteristic extraction method implemented through considering sparsity of multilinear group in speech | |
CN111209853B (en) | Optical fiber sensing vibration signal mode identification method based on AdaBoost-ESN algorithm | |
CN112985574A (en) | High-precision classification identification method for optical fiber distributed acoustic sensing signals based on model fusion | |
CN103065192A (en) | Volume tense data modeling method based on determined learning theory | |
CN110533094A (en) | A kind of evaluation method and system for driver | |
CN104964736B (en) | Optical fiber invasion vibration source identification method based on time-frequency characteristic maximum expected classification | |
Shi et al. | Φ-OTDR Event Recognition System Based on Valuable Data Selection | |
Moustris et al. | Wind power forecasting using historical data and artificial neural networks modeling | |
CN205003281U (en) | Unmanned vehicles early warning device |
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 |