CN106197646A - The detection of a kind of fiber-optic vibration reduces the method for error and fine vibration detection device - Google Patents
The detection of a kind of fiber-optic vibration reduces the method for error and fine vibration detection device Download PDFInfo
- Publication number
- CN106197646A CN106197646A CN201610472547.0A CN201610472547A CN106197646A CN 106197646 A CN106197646 A CN 106197646A CN 201610472547 A CN201610472547 A CN 201610472547A CN 106197646 A CN106197646 A CN 106197646A
- Authority
- CN
- China
- Prior art keywords
- vibration
- rank
- signal
- weight coefficient
- test
- 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)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses the detection of a kind of fiber-optic vibration and reduce the method for error and fine vibration detection device, the method includes: obtain the characteristic group of sample vibration signal；It is weighted processing to the characteristic group of sample vibration signal, obtains test vibration rank, and test vibration rank is mated with the expectation vibration rank of sample vibration signal；When test vibration rank is mated with the expectation vibration rank of sample vibration signal, determine that current weight coefficient is for setting weight coefficient；When test vibration rank is not mated with the expectation vibration rank of sample vibration signal, adjust current weight coefficient, and use the execution of the weight coefficient after adjustment to be weighted processing to the characteristic group of sample vibration signal, obtain test vibration rank, and test vibration rank is carried out the step mated, until test vibration rank is mated with expectation vibration rank with the expectation vibration rank of sample vibration signal.By the way, it is possible to increase the accuracy of fiber-optic vibration detection, the probability of false-alarm is reduced.
Description
Technical field
The present invention relates to vibration detection technology field, particularly relate to a kind of fiber-optic vibration detection reduce error method and
Fine vibration detection device.
Background technology
Using optical fiber can realize distance detection for no reason at all as sensor, optical fiber, along certain path profile, utilizes
Optical fiber internal stress under ambient pressure, mechanical vibration changes the signal disturbance perception brought, or the invasion of outer bound pair network
Realize location and vibration mode identification.Provide an advantage in that spatial distribution and the time simultaneously obtaining measured vibration signal
Change.Have on-the-spot passive, anticorrosive with electromagnetic interference, highly sensitive.
But high sensitivity also results in rate of false alarm height, now uses thresholding more, the mode of pattern recognition filters false-alarm.One, door
Mentality of designing in limit mainly filters from source, extracts the amplitude characteristic of signal, also includes clutter and noise amplitude simultaneously
Change.The amplitude characteristic extracted is compared with thresholding, reports to the police after will be above the signal extraction of thresholding.Two, pattern recognition is main
Utilize CFAR (Constant False-Alarm Rate, constant false alarm rate) algorithm by extracting vibration signal amplitude, time, frequently
The features such as rate carry out Classification and Identification judgement.
The subject matter that existing research exists: use gate method can only filter for single-point, have ignored this point
The Vibration Condition of periphery vibration signal, and only filter on amplitude characteristic, lack considering of frequency, traffic information etc.,
Therefore use the mode bottleneck of thresholding greatly, even if being the most also merely able to distinguish whether single-point vibrates, it is impossible to comprehensive descision
Vibration source classification, the extent of injury, causes false-alarm quantity big.And use and comprehensively can examine on space-time on the theoretical method of pattern recognition
Amount, but owing to the disturbing signal of optical fiber has larger difference with voice signal, therefore do not agree with fiber-optic vibration model.
Summary of the invention
The technical problem that present invention mainly solves is to provide the detection of a kind of fiber-optic vibration and reduces the method for error and fine vibration
Detection device, it is possible to increase the accuracy of fiber-optic vibration detection, reduces the probability of false-alarm.
For solving above-mentioned technical problem, the technical scheme that the present invention uses is: provide a kind of fiber-optic vibration detection to subtract
The method of little error, the method includes: obtain the characteristic group of sample vibration signal；Characteristic to sample vibration signal
Group is weighted processing, and obtains test vibration rank, and by the expectation vibration rank of test vibration rank with sample vibration signal
Mate；When test vibration rank is mated with the expectation vibration rank of sample vibration signal, determine that current weight coefficient is for setting
Determine weight coefficient；When test vibration rank is not mated with the expectation vibration rank of sample vibration signal, adjust current weight coefficient, and
Use the weight coefficient after adjusting to perform the characteristic group of sample vibration signal and be weighted processing, obtain test vibration level
Not, and test vibration rank is carried out with the expectation vibration rank of sample vibration signal the step mated, until test vibration level
Do not mate with expectation vibration rank.
Wherein, it is weighted processing to the characteristic group of sample vibration signal, obtains test vibration rank, including: adopt
With setting weight coefficient, the characteristic group of sample vibration signal is weighted summation, obtains test vibration rank.
Wherein, characteristic group include from vibration signal extract in order to represent oscillation intensity, frequency of vibration, vibration time
Between characteristic at least one.
Wherein, characteristic group includes that M element, vibration rank include N number of rank, set the weight coefficient power system as M × N
Array；Wherein M, M are positive integer.
Wherein, when test vibration rank is not mated with the expectation vibration rank of sample vibration signal, adjust and currently weigh system
Number, including: when the expectation vibration rank of test vibration rank and sample vibration signal is not mated, according to test vibration rank and
The other error size of expectation vibration level of sample vibration signal, is adjusted correspondingly current weight coefficient.
Wherein, the method also includes: obtain the characteristic group of the vibration signal of target fiber position；Use and set power system
Several characteristic groups to sample vibration signal are weighted processing, to obtain the vibration rank of target fiber position.
Wherein, obtain the characteristic group of the vibration signal of target fiber position, including: it is passed through the first light letter to optical fiber
Number, receive first optical signal the second optical signal via fiber reflection；Second optical signal is converted into the signal of telecommunication；Extract the signal of telecommunication
Multiple features, using multiple features as the characteristic group of vibration signal.
For solving above-mentioned technical problem, another technical solution used in the present invention is: provide a kind of fiber-optic vibration detection
Device, this device includes: acquisition module, for obtaining the characteristic group of sample vibration signal；Processing module, for sample
The characteristic group of vibration signal is weighted processing, and obtains test vibration rank, and test vibration rank is vibrated with sample
The expectation vibration rank of signal is mated；Determine module, shake for the expectation in test vibration rank with sample vibration signal
During dynamic rank coupling, determine that current weight coefficient is for setting weight coefficient；Adjusting module, for vibrating with sample in test vibration rank
When the expectation vibration rank of signal is not mated, adjust current weight coefficient, and use the weight coefficient after adjustment to perform sample vibration
The characteristic group of signal is weighted processing, and obtains test vibration rank, and by test vibration rank and sample vibration signal
Expectation vibration rank carry out the step mated, mate until test vibration rank vibrates rank with expectation.
Wherein, processing module is specifically for using setting weight coefficient to be weighted the characteristic group of sample vibration signal
Summation, obtains test vibration rank.
Wherein, characteristic group include from vibration signal extract in order to represent oscillation intensity, frequency of vibration, vibration time
Between characteristic at least one.
The invention has the beneficial effects as follows: be different from the situation of prior art, the fiber-optic vibration detection of the present invention reduces error
Method include: obtain sample vibration signal characteristic group；The characteristic group of sample vibration signal is weighted place
Reason, obtains test vibration rank, and test vibration rank is mated with the expectation vibration rank of sample vibration signal；Surveying
When examination vibration rank is mated with the expectation vibration rank of sample vibration signal, determine that current weight coefficient is for setting weight coefficient；Surveying
When the expectation vibration rank of examination vibration rank and sample vibration signal is not mated, adjust current weight coefficient, and after using adjustment
Weight coefficient performs to be weighted the characteristic group of sample vibration signal processing, and obtains test vibration rank, and test is shaken
Dynamic rank carries out the step mated with the expectation vibration rank of sample vibration signal, until test vibration rank and expectation vibration level
Do not mate.By the way, it is possible to by network neural algorithm, the characteristic group of vibration signal is being converted into vibration level
During other, use the error of reality output and desired output that weight coefficient is modified so that actual output is more nearly
Desired output, thus in the detection of fiber-optic vibration rank, improve the accuracy of fiber-optic vibration detection, reduce the probability of false-alarm.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet that fiber-optic vibration of the present invention detection reduces method one embodiment of error；
Fig. 2 is the structural framing schematic diagram that fiber-optic vibration of the present invention detection reduces method one embodiment of error；
Fig. 3 is that fiber-optic vibration of the present invention detects two-layer neural network structure signal in method one embodiment reducing error
Figure；
Fig. 4 is that fiber-optic vibration of the present invention detects multilayer neural network structural representation in method one embodiment reducing error
Figure；
Fig. 5 is the structural representation of fiber-optic vibration of the present invention detection device one embodiment；
Fig. 6 is the structural representation of fiber-optic vibration of the present invention detection another embodiment of device.
Detailed description of the invention
Below describe in, in order to illustrate rather than in order to limit, it is proposed that such as particular system structure, interface, technology it
The detail of class, in order to thoroughly understand the application.But, it will be clear to one skilled in the art that and do not having these concrete
Other embodiment of details can also realize the application.In other situation, omit to well-known device, circuit with
And the detailed description of method, in order to avoid unnecessary details hinders the description of the present application.
It is the schematic flow sheet that fiber-optic vibration of the present invention detection reduces method one embodiment of error refering to Fig. 1, Fig. 1,
The method includes:
S11: obtain the characteristic group of sample vibration signal.
Wherein, sample vibration signal refers to the vibration signal tested, and this sample vibration signal can be artificial at light
The vibration that fine predetermined position causes, therefore, the vibration rank of this sample vibration signal is known, i.e. mention in S12
The expectation vibration rank of sample vibration signal.
For example, it is possible to make a heavy truck rolling road surface, the fiber laser arrays arranged under this road surface is to vibration signal, if people
The vibration rank of the heavy truck rolling road surface for arranging is 1 grade, then the vibration rank that this vibration signal is corresponding is also 1 grade.
Optionally, characteristic group include from vibration signal extract in order to represent oscillation intensity, frequency of vibration, vibration
At least one in the characteristic of time.
S12: be weighted processing to the characteristic group of sample vibration signal, obtains test vibration rank, and will test
Vibration rank is vibrated rank with the expectation of sample vibration signal and is mated.
Optionally, in one embodiment, the mode that weighting processes is weighted sum, i.e. uses and sets weight coefficient to sample
The characteristic group of this vibration signal is weighted summation, obtains test vibration rank.
Wherein, this setting weight coefficient can be default one be not the numeral of 0, the numeral between usually 0-1, weighting
The mode of summation can be that each characteristic is multiplied by this weight coefficient, more multiple characteristics are multiplied by the result of weight coefficient
It is added summation, thus obtains a numerical value, by numerical value and the other corresponding relation of vibration level, it is possible to obtain this vibration signal
Test vibration rank.
Such as, characteristic is [X1, X2, X3], wherein, X1、X2、X3It is respectively the amplitude of sample vibration signal, frequency, shakes
Dynamic time and system peak swing, frequency, the ratio of time of vibration, therefore, X1、X2、X3For the numerical value between [0,1], then use
The weight coefficient W numerical value obtained that is weighted this feature data set suing for peace is: W X1+W X2+W X3.It is appreciated that summation
Result should be [0,3], therefore, it can to its be estimated arrange, such as result be [0,1.5), vibration rank be 1 grade, tie
Be really [1.5,2.5), vibration rank is 2 grades, and result is [2.5,3], and vibration rank is 3 grades.
Optionally, it is also possible to for different characteristics, different weight coefficients, such as, the spy of sample vibration signal are set
Levying data set is Xi=[X1, X2, X3], weight coefficient Wi=[W1, W2, W3], then the result of weighted sum is: W1X1+W2X2+W3X3。
Optionally, characteristic group includes that M element, vibration rank include N number of rank, set the weight coefficient power as M × N
Coefficient sets；Wherein M, M are positive integer.
Specifically, weight coefficient group can be a matrix, such as:
S13: when test vibration rank is mated with the expectation vibration rank of sample vibration signal, determine that current weight coefficient is
Set weight coefficient.
In conjunction with above for the vibration example that determines of rank, the i.e. result of weighted sum be [0,1.5), vibrating rank is 1
Level, result be [1.5,2.5), vibration rank be 2 grades, result is [2.5,3], vibrate rank be 3 grades.If sample vibration signal is special
The result levying data set weighted sum is 1.8, then test vibration rank should be 2 grades, and the expectation that sample vibration signal is corresponding
Vibration rank is also 2 grades, then it is believed that test vibration rank is mated with the expectation vibration rank of sample vibration signal, and will weighting
The weight coefficient W used in summation process is as the weight coefficient set.
S14: when test vibration rank is not mated with the expectation vibration rank of sample vibration signal, adjust and currently weigh system
Number.
After S14 completes, the weight coefficient after adjusting is used again to perform S12, until test vibration rank and expectation vibration
Rank is mated.
Optionally, S14 can be specifically: vibrates rank in test vibration rank with the expectation of sample vibration signal and does not mates
Time, according to the other error size of expectation vibration level of test vibration rank Yu sample vibration signal, current weight coefficient is carried out phase
The adjustment answered.
Such as, when test vibration is superior to the expectation vibration rank of vibration signal, weight coefficient suitably can be turned down,
When test vibration rank is less than the expectation vibration rank of vibration signal, weight coefficient suitably can be tuned up.Wherein by weight coefficient
Tune up the degree turned down and can be superior to the other error size of expectation vibration level of vibration signal according to test vibration, the most permissible
It is manually set, it is also possible to be system default.
Optionally, after S13 or S14, if weight coefficient determines, it is possible to implement vibration detection by this weight coefficient
, i.e. obtain the characteristic group of the vibration signal of target fiber position；Use and set the weight coefficient spy to sample vibration signal
Levy data set to be weighted processing, to obtain the vibration rank of target fiber position.
Concrete, obtain the characteristic group of the vibration signal of target fiber position, specifically may include that
Obtain the characteristic group of the vibration signal of target fiber position, including:
It is passed through the first optical signal to optical fiber, receives first optical signal the second optical signal via fiber reflection；By the second light
Signal is converted into the signal of telecommunication；Extract multiple features of the signal of telecommunication, using multiple features as the characteristic group of vibration signal.
Specifically, can be passed through the first optical signal of modulation to optical fiber, this first optical signal is pulsed optical signals, first
After optical signal is passed through optical fiber, connecing of the first optical signal can be passed through in the generation Rayleigh scattering everywhere (or backscattering) of optical fiber
The second optical signal of reflection can be received at Kou, thus detect multiple concurrent vibration source simultaneously, thus realize early warning and to shaking
Source location.
After receiving the second optical signal, the second optical signal is sampled, obtain multiple corresponding different fiber position
Optical signal.Wherein, the distance between fiber position and optical fiber connector (being i.e. passed through the interface of the first optical signal) can be by as follows
Formula calculates:
Wherein, L is the distance between fiber position and optical fiber connector, and c is the light velocity, and T is for being passed through the first optical signal to receiving
The duration of the second optical signal, n is the refractive index of optical fiber.
The signal of telecommunication that the optical signal that sampling obtains is converted to correspondence the most again is easy to the process of signal.Here can pass through
General optical-electrical converter such as APD is converted to analogue signal, then converts analog signals into numeral letter by analog-digital converter
Number.
In other embodiments, first the second optical signal can be converted into the signal of telecommunication, then the signal of telecommunication is sampled with
Obtain the sampled signal of different fiber position.
When extracting the characteristic in the signal of telecommunication, can be as follows:
Such as, the maximum vibration amplitude of system default is 100mV, and maximum vibration frequency is 100Hz, and the maximum vibration time is
10 minutes, if the vibration amplitude of the target fiber position vibration signal got is 30mV, frequency of vibration was 40Hz, time of vibration
Be 1 minute, then this feature data set can be [0.3,0.4,0.1], i.e. each data and the ratio of maximum data.
Specifically, as in figure 2 it is shown, Fig. 2 be fiber-optic vibration of the present invention detection reduce error method one embodiment be
System block schematic illustration.
Above-mentioned embodiment can use the system framework such as Fig. 2 to describe, and input unit is used for inputting sample vibration signal
Characteristic group X, training department is used for using weight coefficient W to be weighted characteristic group processing, and output unit adds for output
Result Y that power processes, then exports reality result Y and feeds back to training department, and it is defeated with expectation that reality is exported result Y by training department again
Go out result Y0Compare, form error e, by error e, weight coefficient W is modified, so that the result of test output next time
Desired output result Y can be more nearly0。
In one embodiment, output unit is a Rotating fields, as shown in Figure 3.
Test output Y=W1X1+W2X2+…+WnXn, by test output Y and sample desired output Y0Relatively, obtain error e,
Error e is used in turn weight coefficient W to be adjusted again, so that test output Y is more nearly sample desired output Y0。
In another embodiment, output unit is multiple structure, as shown in Figure 4, similar neutral net, therein each
The similar neuron of point.
Below as a example by multiple structure, present embodiment is described in detail:
First to the weight coefficient W needed each layerijPut initial value.Optionally, can be to the weight coefficient W of each layerijPut one relatively
Little non-zero random number, wherein Wi,n+1=-θ.Wherein, θ is a threshold value set.
Then the characteristic group X=(X of sample vibration signal is obtainedl, X2..., Xn, 1), corresponding desired output Y1, Y1Can
Think vibration rank.This vibration rank is rule of thumb manually to be judged corresponding alarm situation by engineering staff, and by
This generates the vibration rank of corresponding alarm.
Calculate the output of each layer the most again, wherein, for the input X of the i-th neuron of kth layerik, have:
Wherein, Xn+1,k-1=1, Wi,n+1=-θ.
Then at the learning error e calculating each layerij, for output layer kth layer neuron, have:
eik=Yik(1-Yik)(Yi-Yik)
Wherein, YiFor desired output.
For the neuron j and neuron i before of other before output layer layer, have:
Finally, modified weight coefficient Wij:
Wij(t+1)=Wij(t)+ηejYi
Wherein, t is the number of times revised, 0 < η < 1.
After having obtained each weight coefficient of each layer, the characteristic group of sample vibration signal just can be used again to carry out
Test, if for given sample Xp=(Xp1, Xp2... Xpn, 1) and desired output Yp=(Yp1) all meet input and output requirement,
Then training terminates.
Being different from prior art, the fiber-optic vibration detection of present embodiment reduces the method for error and includes: obtains sample and shakes
The characteristic group of dynamic signal；It is weighted processing to the characteristic group of sample vibration signal, obtains test vibration rank, and
With the expectation of sample vibration signal, test vibration rank is vibrated rank mate；At test vibration rank and sample vibration letter
Number expectation vibration rank coupling time, determine current weight coefficient for set weight coefficient；At test vibration rank and sample vibration letter
Number expectation vibration rank when not mating, adjust current weight coefficient, and use the weight coefficient after adjustment to perform sample vibration letter
Number characteristic group be weighted processing, obtain test vibration rank, and by test vibration rank and sample vibration signal
Expect to vibrate the step that rank carries out mating, until test vibration rank is mated with expectation vibration rank.By the way, energy
Enough by network neural algorithm the characteristic group of vibration signal is converted into vibration level other during, use reality output
With the error of desired output, weight coefficient is modified so that actual output is more nearly desired output, thus at fiber-optic vibration
In the detection of rank, improve the accuracy of fiber-optic vibration detection, reduce the probability of false-alarm.
Being the structural representation of fiber-optic vibration of the present invention detection device one embodiment refering to Fig. 5, Fig. 5, this device includes:
Acquisition module 51, for obtaining the characteristic group of sample vibration signal.
Optionally, characteristic group include from vibration signal extract in order to represent oscillation intensity, frequency of vibration, vibration
At least one in the characteristic of time.
Such as, weight coefficient Wi=[W1, W2, W3]。
Optionally, characteristic group includes that M element, vibration rank include N number of rank, set the weight coefficient power as M × N
Coefficient sets；Wherein M, M are positive integer.
Specifically, weight coefficient group can be a matrix, such as:
Processing module 52, for being weighted processing to the characteristic group of sample vibration signal, obtains test vibration level
Not, and with the expectation of sample vibration signal, test vibration rank is vibrated rank to mate.
Optionally, processing module 52 is specifically for using setting weight coefficient to carry out the characteristic group of sample vibration signal
Weighted sum, obtains test vibration rank.
Such as, the characteristic group of sample vibration signal is Xi=[X1, X2, X3], weight coefficient Wi=[W1, W2, W3], then add
The result of power summation is: W1X1+W2X2+W3X3。
Determine module 53, for when test vibration rank is mated with the expectation vibration rank of sample vibration signal, determining
Current weight coefficient is for setting weight coefficient.
Adjusting module 54, for when test vibration rank is not mated with the expectation vibration rank of sample vibration signal, adjusting
Whole current weight coefficient, and use the execution of the weight coefficient after adjustment to be weighted processing to the characteristic group of sample vibration signal,
Obtain test vibration rank, and test vibration rank carried out with the expectation vibration rank of sample vibration signal the step mated,
Until test vibration rank is mated with expectation vibration rank.
Optionally, adjusting module 54 is specifically for vibrating rank not in the expectation of test vibration rank with sample vibration signal
During coupling, according to the other error size of expectation vibration level of test vibration rank Yu sample vibration signal, current weight coefficient is entered
Row is corresponding to be adjusted.
Optionally, acquisition module 51 is additionally operable to obtain the characteristic group of the vibration signal of target fiber position.Process mould
Block 52 is additionally operable to use setting weight coefficient to be weighted the characteristic group of sample vibration signal processing, so that determining module 53
Obtain the vibration rank of target fiber position.
Optionally, this device also includes acquisition module, for being passed through the first optical signal to optical fiber, receives the first optical signal warp
By the second optical signal of fiber reflection；Second optical signal is converted into the signal of telecommunication；Extract multiple features of the signal of telecommunication, by multiple spies
Levy the characteristic group as vibration signal.
It should be understood that the device that present embodiment provides is method based on above-mentioned embodiment and proposes, in fact
The principle executed is similar with step, repeats no more here.
It is the structural representation of fiber-optic vibration of the present invention detection another embodiment of device refering to Fig. 6, Fig. 6, this device bag
Include processor 61, memorizer 62, receptor 63 and transmitter 64.
Wherein, processor 61, memorizer 62, receptor 63 and transmitter 64 may each be one or more.
Receptor 63 is for receiving the information that external equipment sends, for example, it may be the optical fiber that Fibre Optical Sensor sends shakes
Dynamic signal or the characteristic group therefrom extracted.
Transmitter 64 is for sending the result of process, for example, it is possible to alarm signal is sent to alarm, display
Device etc., certainly, in other embodiments, it is also possible to there is no transmitter 64.
Memorizer 62 is for storing system file, application software and default algorithm, threshold value, it is also possible to storing fiber optic is gone through
History vibration signal or record other to vibration level.Wherein, memorizer 62 can include read only memory, random access memory and
At least one in nonvolatile RAM (NVRAM).
Processor 61 is used for performing following steps:
The characteristic group of sample vibration signal is obtained by receptor 63.
It is weighted processing to the characteristic group of sample vibration signal, obtains test vibration rank, and by test vibration
Rank is vibrated rank with the expectation of sample vibration signal and is mated.
When test vibration rank is mated with the expectation vibration rank of sample vibration signal, determine that current weight coefficient is for setting
Weight coefficient.
When test vibration rank is not mated with the expectation vibration rank of sample vibration signal, adjust current weight coefficient, and
Use the weight coefficient after adjusting to perform the characteristic group of sample vibration signal and be weighted processing, obtain test vibration level
Not, and test vibration rank is carried out with the expectation vibration rank of sample vibration signal the step mated, until test vibration level
Do not mate with expectation vibration rank.
Optionally, processor 61 is specifically for using setting weight coefficient to add the characteristic group of sample vibration signal
Power summation, obtains test vibration rank.
Wherein, characteristic group include from vibration signal extract in order to represent oscillation intensity, frequency of vibration, vibration time
Between characteristic at least one.
Wherein, characteristic group includes that M element, vibration rank include N number of rank, set the weight coefficient power system as M × N
Array；Wherein M, M are positive integer.
Optionally, processor 61 is additionally operable to not mate with the expectation vibration rank of sample vibration signal in test vibration rank
Time, according to the other error size of expectation vibration level of test vibration rank Yu sample vibration signal, current weight coefficient is carried out phase
The adjustment answered.
Optionally, processor 61 is additionally operable to obtain the characteristic group of the vibration signal of target fiber position；Use and set
The characteristic group of sample vibration signal is weighted processing, to obtain the vibration rank of target fiber position by weight coefficient.
Wherein, processor 61 is specifically for being passed through the first optical signal by transmitter 64 to optical fiber, then passes through receptor 63
Receive first optical signal the second optical signal via fiber reflection；Second optical signal is converted into the signal of telecommunication；Extract the signal of telecommunication
Multiple features, using multiple features as the characteristic group of vibration signal.
Above-mentioned processor 61 can also be referred to as CPU (Central Processing Unit, CPU).Specifically
Application in, each assembly of terminal is coupled by bus, and wherein bus is in addition to including data/address bus, it is also possible to bag
Include power bus, control bus and status signal bus in addition etc..But for the sake of understanding explanation, in the drawings various buses are all marked
For bus.
It should be understood that processor 61 is probably a kind of IC chip, there is the disposal ability of signal.Realizing
Cheng Zhong, each step of said method can be by the integrated logic circuit of the hardware in processor 61 or the instruction of software form
Complete.Above-mentioned processor 61 can be general processor, digital signal processor (DSP), special IC (ASIC), show
Become programmable gate array (FPGA) or other PLDs, discrete gate or transistor logic, discrete hardware
Assembly.Can realize or perform disclosed each method, step and the logic diagram in the embodiment of the present invention.General processor can
To be microprocessor or this processor can also be the processor etc. of any routine.In conjunction with the side disclosed in the embodiment of the present invention
The step of method can be embodied directly in hardware decoding processor and perform, or with the hardware in decoding processor and software mould
Block combination execution completes.Software module may be located at random access memory, flash memory, read only memory, programmable read only memory or
In the storage medium that this areas such as person's electrically erasable programmable memorizer, depositor are ripe.This storage medium is positioned at memorizer 62,
Processor 61 reads the information in respective memory, completes the step of said method in conjunction with its hardware.
In several embodiments provided by the present invention, it should be understood that disclosed method and equipment, Ke Yitong
The mode crossing other realizes.Such as, equipment embodiment described above is only schematically, such as, described module or
The division of unit, is only a kind of logic function and divides, and actual can have other dividing mode, the most multiple unit when realizing
Or assembly can in conjunction with or be desirably integrated into another system, or some features can be ignored, or does not performs.
The described unit illustrated as separating component can be or may not be physically separate, shows as unit
The parts shown can be or may not be physical location, i.e. may be located at a place, or can also be distributed to multiple
On NE.Some or all of unit therein can be selected according to the actual needs to realize present embodiment scheme
Purpose.
It addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it is possible to
Being that unit is individually physically present, it is also possible to two or more unit are integrated in a unit.Above-mentioned integrated
Unit both can realize to use the form of hardware, it would however also be possible to employ the form of SFU software functional unit realizes.
If the integrated unit in other embodiments above-mentioned realizes and as independent using the form of SFU software functional unit
Production marketing or use time, can be stored in a computer read/write memory medium.Based on such understanding, the present invention
The part that the most in other words prior art contributed of technical scheme or this technical scheme the most permissible
Embodying with the form of software product, this computer software product is stored in a storage medium, uses including some instructions
So that a computer equipment (can be personal computer, server, or the network equipment etc.) or processor
(processor) all or part of step of method described in each embodiment of the present invention is performed.And aforesaid storage medium bag
Include: USB flash disk, portable hard drive, read only memory (ROM, Read-Only Memory), random access memory (RAM, Random
Access Memory), the various media that can store program code such as magnetic disc or CD.
The foregoing is only embodiments of the present invention, not thereby limit the scope of the claims of the present invention, every utilization is originally
Equivalent structure or equivalence flow process that description of the invention and accompanying drawing content are made convert, or are directly or indirectly used in what other were correlated with
Technical field, is the most in like manner included in the scope of patent protection of the present invention.
Claims (10)
1. the method that a fiber-optic vibration detection reduces error, it is characterised in that including:
Obtain the characteristic group of sample vibration signal；
It is weighted processing to the characteristic group of described sample vibration signal, obtains test vibration rank, and by described test
Vibration rank is vibrated rank with the expectation of described sample vibration signal and is mated；
When described test vibration rank is mated with the expectation vibration rank of described sample vibration signal, determine that current weight coefficient is
Set weight coefficient；
When described test vibration rank is not mated with the expectation vibration rank of described sample vibration signal, adjust and currently weigh system
Number, and use the described characteristic group to described sample vibration signal of the execution of the weight coefficient after adjustment to be weighted processing,
To test vibration rank, and carry out described test vibration rank with the expectation vibration rank of described sample vibration signal mating
Step, until described test vibration rank is mated with described expectation vibration rank.
Method the most according to claim 1, it is characterised in that
The described characteristic group to described sample vibration signal is weighted processing, and obtains test vibration rank, including:
Use described setting weight coefficient that the characteristic group of described sample vibration signal is weighted summation, obtain test vibration
Rank.
Method the most according to claim 2, it is characterised in that
Described characteristic group include extracting from vibration signal in order to represent oscillation intensity, frequency of vibration, time of vibration
At least one in characteristic.
Method the most according to claim 2, it is characterised in that
Described characteristic group includes that M element, described vibration rank include N number of rank, and the described weight coefficient that sets is as M × N's
Weight coefficient group；
Wherein M, M are positive integer.
Method the most according to claim 1, it is characterised in that
Described when described test vibration rank is not mated with the expectation vibration rank of described sample vibration signal, adjust current power
Coefficient, including:
When described test vibration rank is not mated with the expectation vibration rank of described sample vibration signal, shake according to described test
Dynamic rank and the other error size of expectation vibration level of described sample vibration signal, be adjusted correspondingly current weight coefficient.
Method the most according to claim 1, it is characterised in that described method also includes:
Obtain the characteristic group of the vibration signal of target fiber position；
Described setting weight coefficient is used to be weighted the characteristic group of described sample vibration signal processing, to obtain described mesh
The vibration rank of mark fiber position.
Method the most according to claim 6, it is characterised in that
The characteristic group of the vibration signal of described acquisition target fiber position, including:
It is passed through the first optical signal to optical fiber, receives described first optical signal the second optical signal via described fiber reflection；
Described second optical signal is converted into the signal of telecommunication；
Extract multiple features of the described signal of telecommunication, using the plurality of feature as the characteristic group of vibration signal.
8. a fiber-optic vibration detection device, it is characterised in that including:
Acquisition module, for obtaining the characteristic group of sample vibration signal；
Processing module, for being weighted processing to the characteristic group of described sample vibration signal, obtains test vibration rank,
And described test vibration rank is mated with the expectation vibration rank of described sample vibration signal；
Determine module, be used for when described test vibration rank is mated with the expectation vibration rank of described sample vibration signal, really
Settled front weight coefficient is for setting weight coefficient；
Adjusting module, is used for when described test vibration rank is not mated with the expectation vibration rank of described sample vibration signal,
Adjust current weight coefficient, and use the described characteristic group to described sample vibration signal of the execution of the weight coefficient after adjustment to carry out
Weighting processes, and obtains test vibration rank, and by the expectation vibration level of described test vibration rank Yu described sample vibration signal
Do not carry out the step mated, until described test vibration rank is mated with described expectation vibration rank.
Device the most according to claim 8, it is characterised in that
The characteristic group of described sample vibration signal is carried out by described processing module specifically for using described setting weight coefficient
Weighted sum, obtains test vibration rank.
Device the most according to claim 9, it is characterised in that
Described characteristic group include extracting from vibration signal in order to represent oscillation intensity, frequency of vibration, time of vibration
At least one in characteristic.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610472547.0A CN106197646A (en) | 2016-06-24 | 2016-06-24 | The detection of a kind of fiber-optic vibration reduces the method for error and fine vibration detection device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610472547.0A CN106197646A (en) | 2016-06-24 | 2016-06-24 | The detection of a kind of fiber-optic vibration reduces the method for error and fine vibration detection device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106197646A true CN106197646A (en) | 2016-12-07 |
Family
ID=57461678
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610472547.0A Pending CN106197646A (en) | 2016-06-24 | 2016-06-24 | The detection of a kind of fiber-optic vibration reduces the method for error and fine vibration detection device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106197646A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109325402A (en) * | 2018-08-06 | 2019-02-12 | 高维度（深圳）生物信息智能应用有限公司 | A kind of signal processing method, system and computer storage medium |
| CN111951505A (en) * | 2020-08-25 | 2020-11-17 | 青岛大学 | Fence vibration intrusion location and pattern recognition method based on distributed optical fiber system |
| CN112562237A (en) * | 2020-12-11 | 2021-03-26 | 无锡科晟光子科技有限公司 | Optical fiber vibration system positioning accuracy optimization method applied to fence |
| WO2023119806A1 (en) * | 2021-12-24 | 2023-06-29 | 日本電信電話株式会社 | Determining device, determining method, and program |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060122809A1 (en) * | 2004-12-06 | 2006-06-08 | Clarke Burton R | Vibration analysis system and method for a machine |
| CN102354422A (en) * | 2011-10-19 | 2012-02-15 | 湖南德顺电子科技有限公司 | Perimeter protection-oriented method for monitoring suspicious target |
| CN103558042A (en) * | 2013-10-28 | 2014-02-05 | 中国石油化工股份有限公司 | Rapid unit failure diagnosis method based on full state information |
| CN103617684A (en) * | 2013-12-12 | 2014-03-05 | 威海北洋电气集团股份有限公司 | Interference type optical fiber perimeter vibration intrusion recognition algorithm |
| CN104964736A (en) * | 2015-07-15 | 2015-10-07 | 北方工业大学 | Optical fiber invasion vibration source identification method based on time-frequency characteristic EM classification |
| CN105046859A (en) * | 2015-06-02 | 2015-11-11 | 北方工业大学 | Optical fiber intrusion detection method based on vibration signal space-time two-dimensional sparse representation K-S detection |
| CN105067101A (en) * | 2015-08-05 | 2015-11-18 | 北方工业大学 | Fundamental tone frequency characteristic extraction method based on vibration signal for vibration source identification |
| CN105222885A (en) * | 2015-06-19 | 2016-01-06 | 北方工业大学 | Optical fiber vibration detection method and device |
| CN105606198A (en) * | 2016-01-28 | 2016-05-25 | 吉林大学 | Fiber vibration sensing system two-order signal feature extraction and determining method |
-
2016
- 2016-06-24 CN CN201610472547.0A patent/CN106197646A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060122809A1 (en) * | 2004-12-06 | 2006-06-08 | Clarke Burton R | Vibration analysis system and method for a machine |
| CN102354422A (en) * | 2011-10-19 | 2012-02-15 | 湖南德顺电子科技有限公司 | Perimeter protection-oriented method for monitoring suspicious target |
| CN103558042A (en) * | 2013-10-28 | 2014-02-05 | 中国石油化工股份有限公司 | Rapid unit failure diagnosis method based on full state information |
| CN103617684A (en) * | 2013-12-12 | 2014-03-05 | 威海北洋电气集团股份有限公司 | Interference type optical fiber perimeter vibration intrusion recognition algorithm |
| CN105046859A (en) * | 2015-06-02 | 2015-11-11 | 北方工业大学 | Optical fiber intrusion detection method based on vibration signal space-time two-dimensional sparse representation K-S detection |
| CN105222885A (en) * | 2015-06-19 | 2016-01-06 | 北方工业大学 | Optical fiber vibration detection method and device |
| CN104964736A (en) * | 2015-07-15 | 2015-10-07 | 北方工业大学 | Optical fiber invasion vibration source identification method based on time-frequency characteristic EM classification |
| CN105067101A (en) * | 2015-08-05 | 2015-11-18 | 北方工业大学 | Fundamental tone frequency characteristic extraction method based on vibration signal for vibration source identification |
| CN105606198A (en) * | 2016-01-28 | 2016-05-25 | 吉林大学 | Fiber vibration sensing system two-order signal feature extraction and determining method |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109325402A (en) * | 2018-08-06 | 2019-02-12 | 高维度（深圳）生物信息智能应用有限公司 | A kind of signal processing method, system and computer storage medium |
| CN109325402B (en) * | 2018-08-06 | 2022-04-15 | 高维度（深圳）生物信息智能应用有限公司 | Signal processing method, system and computer storage medium |
| CN111951505A (en) * | 2020-08-25 | 2020-11-17 | 青岛大学 | Fence vibration intrusion location and pattern recognition method based on distributed optical fiber system |
| CN111951505B (en) * | 2020-08-25 | 2022-02-08 | 青岛大学 | Fence vibration intrusion positioning and mode identification method based on distributed optical fiber system |
| CN112562237A (en) * | 2020-12-11 | 2021-03-26 | 无锡科晟光子科技有限公司 | Optical fiber vibration system positioning accuracy optimization method applied to fence |
| WO2023119806A1 (en) * | 2021-12-24 | 2023-06-29 | 日本電信電話株式会社 | Determining device, determining method, and program |
| WO2023119628A1 (en) * | 2021-12-24 | 2023-06-29 | 日本電信電話株式会社 | Determining device, determining method, and program |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104751580B (en) | The recognition methods of distributing optical fiber sensing signal mode and system | |
| Wu et al. | One-dimensional CNN-based intelligent recognition of vibrations in pipeline monitoring with DAS | |
| Li et al. | Damage identification in civil engineering structures utilizing PCA‐compressed residual frequency response functions and neural network ensembles | |
| Hernández et al. | Deep-learning-based earthquake detection for fiber-optic distributed acoustic sensing | |
| Shiloh et al. | Efficient processing of distributed acoustic sensing data using a deep learning approach | |
| Shao et al. | Data-driven distributed optical vibration sensors: A review | |
| CN103617684B (en) | Interference-type optical fiber circumference vibrating intruding recognizer | |
| CN106197646A (en) | The detection of a kind of fiber-optic vibration reduces the method for error and fine vibration detection device | |
| US20200104224A1 (en) | Anomaly detection device, anomaly detection method and storage medium | |
| CN110751108A (en) | A Method for Determining Similarity of Subway Distributed Vibration Signals | |
| CN108132092A (en) | Threshold value optical fiber vibration event recognition methods is gone based on adaptive mean value | |
| Chen et al. | Artificial immune pattern recognition for structure damage classification | |
| CN110487391A (en) | Intelligent optical fiber distribution acoustic wave sensing system and method based on AI chip | |
| Jia et al. | Automatic event detection in low SNR microseismic signals based on multi-scale permutation entropy and a support vector machine | |
| CN105973451A (en) | Optical fiber vibration model determination method and device | |
| CN105606198A (en) | Fiber vibration sensing system two-order signal feature extraction and determining method | |
| CN108760305A (en) | A kind of Bearing Fault Detection Method, device and equipment | |
| CN107862785A (en) | Bill authentication method and device | |
| CN108599863A (en) | Overhead transmission line wind based on shallow-layer CNN disaggregated models waves monitoring and pre-warning system and method | |
| CN106301575B (en) | The sorting technique and device and optical fiber sensing system of a kind of fiber-optic vibration signal | |
| CN117195750A (en) | Landslide disaster sensitivity model construction method based on time sequence deformation | |
| Reyes-Galaviz et al. | Analysis of an infant cry recognizer for the early identification of pathologies | |
| CN111951505B (en) | Fence vibration intrusion positioning and mode identification method based on distributed optical fiber system | |
| CN110991507A (en) | Road underground cavity identification method, device and system based on classifier | |
| Duque-Montoya et al. | Automatic ecosystem identification using Psychoacoustical features |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| AD01 | Patent right deemed abandoned | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20191206 |