CN106643484A - Impact positioning method based on transient fractal calculation and similar two-dimensional normal distribution fitting - Google Patents
Impact positioning method based on transient fractal calculation and similar two-dimensional normal distribution fitting Download PDFInfo
- Publication number
- CN106643484A CN106643484A CN201610902301.2A CN201610902301A CN106643484A CN 106643484 A CN106643484 A CN 106643484A CN 201610902301 A CN201610902301 A CN 201610902301A CN 106643484 A CN106643484 A CN 106643484A
- Authority
- CN
- China
- Prior art keywords
- optical fiber
- sensor
- shock
- impact
- coordinate
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/002—Measuring arrangements characterised by the use of optical techniques for measuring two or more coordinates
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention provides an impact positioning method based on transient fractal calculation and similar two-dimensional normal distribution fitting. The method comprises steps of 1, cellular structure monitoring network distribution and corresponding monitoring region division; 2, impact sample point selection rule and impact response signal acquisition; 3, impact response characteristic information extraction based on transient fractal calculation; 4, fiber FBG sensor impact response characteristic curved surface fitting based on a similar two-dimensional normal distribution model; and 5, instantaneous impact load position identification based on fiber FBG sensor impact response characteristic curved surface solution.
Description
Technical field
The invention belongs to the Impact monitoring field of monitoring structural health conditions, specifically propose it is a kind of based on transient state fractal calculation and
The impact load location method of class two-dimensional normal distribution fitting.
Background technology
Early stage im-pact location recognition methods mainly by extracting arrival time of Shock stress Wave, according to distance, the time,
Relation between stress wave velocity of wave three is calculating the positional information of impact.But in a practical situation, the ripple of stress wave reaches
Time extracts easily affected by noise.There is scholar to carry out frequency domain process to signal using the method for wavelet analysis later, reduced
The impact of noise, is extracted ripple up to the time, adjusts the distance, time, Velocity Equation are optimized, and obtain the position letter of impact
Breath.
Although had many good qualities based on the impact load location method of optical fiber FBG, but its sampling rate is relatively low, deposits
In following deficiency.
First, existing optical fiber FBG impact load location method is adopted just mostly in sensor placement mode
Square layout.This layout method can only be applied to large-area sheet structural impact to being monitored in sensor area defined
When in layout effective rate of utilization it is relatively low, therefore be badly in need of research be favorably improved sensor placement monitoring efficiency method.
Secondly, existing Fibre Optical Sensor shock loading location identification extracting method is mostly using FFT, little wavelength-division
Solution processing method, these method for extracting signal all have ignored to a certain extent error produced during sample point signals collecting,
So that result is not accurate enough.And pass through to extract impulse response signal characteristic information, using transcendental method such as neutral net side
The method that method determines the positional information of impact, existing again needs a large amount of different impact conditions to carry out training network, and Generalization Capability is not
Well, the shortcomings of practicality and not strong real-time.Accordingly, it would be desirable to study be favorably improved spectral purity, experimental error is reduced
Method.
Additionally, existing Fibre Optical Sensor characteristic signal processing method great majority can only characterize optical fiber FBG at certain
Feature on one line or in certain part.In order to improve shock loading identification speed, research is needed to characterize optical fiber FBG
Characteristic model of the sensor in monitored area.
Finally, these shock loading location recognition methods are recognized not for time difference method, neutral net, SVMs
Foot, the present invention proposes a kind of Impact Location Method being fitted based on transient state fractal calculation and class two-dimensional normal distribution.
The content of the invention
Technical problem：The invention provides a kind of shock loading position identifying method for being favorably improved efficiency, the method
Using the optical fiber FBG sensing network collocation methods based on hexagonal honeycomb layout, punching is extracted by way of transient state fractal calculation
The characteristic information of response signal is hit, using the high fit characteristic of characteristic information and class two-dimensional normal distribution curved surface curved surface plan is carried out
Close, it is adaptable to large-area thin plate class formation, the method contributes to effectively reducing optical fiber FBG biographies in the case where monitoring area is certain
Sensor quantity, reduces system complexity.
Technical scheme：To solve above-mentioned technical problem, the present invention is intended based on transient state fractal calculation and class two-dimensional normal distribution
The transient load positioning of conjunction, comprises the following steps：
Step one：Honeycomb monitoring network is arranged and corresponding monitored area divides
Monitored area is divided into into seven identical regular hexagon structural regions, and connects into honeycomb shape, i.e., middle one
Individual regular hexagon, six regular hexagons of surrounding；Each one optical fiber FBG of orthohexagonal center lateral arrangement, position
Sensors A is denoted as in middle Fibre Optical Sensor, remaining prolongs clockwise direction and is denoted as sensor B, C, D, E, F, G successively；
Middle regular hexagon is regarded as into six equilateral triangle areas, by each equilateral triangle, and positive six for being adjacent
Side shape is together as a monitored area, and whole monitored area is divided into six monitoring subregions, respectively monitor subregion I,
Ⅱ、Ⅲ、Ⅳ、Ⅴ、Ⅵ；
Wherein sensors A and sensor B monitor subregion I；Sensors A and sensor C monitoring subregions II；Sensors A
Subregion III is monitored with sensor D；Sensors A and sensor E monitoring subregions IV；Sensors A and sensor F monitoring subregions
Ⅴ；Sensors A and sensor G monitoring subregions VI；
Step 2：Impact sample point selection rule is gathered with impulse response signal
(2-1), a two-dimensional Cartesian coordinate system is set up on the impact region to be monitored of structural test piece, with fiber grating biography
Sensor A and E line direction as x-axis positive direction, using fiber-optic grating sensor A and B line direction as x-axis negative direction, with light
Fiber grating sensor A and sensor C, D midpoints line direction as y-axis positive direction, with fiber-optic grating sensor A and sensor F
With G midpoints line direction as y-axis negative direction；Impacting sample point selection rule is：On each two adjacent sensors line
It is even to take a little, and using these points as impact sample point；
(2-2) and then using jump bit these impact sample points are impacted successively using identical energy size；Respectively
Under record is per Secondary Shocks, the impulse response signal of distributing optical fiber sensing network；By the distributing optical fiber sensing under every Secondary Shocks
The impulse response signal of network is referred to as one group of impulse response signal；Every group of impulse response signal includes the corresponding punching of seven sensors
Hit the center wavelength shift amount of the sensor of response signal, i.e., seven；
Step 3：Shock response feature information extraction based on transient state fractal calculation
Successively feature information extraction is carried out to each group of impulse response signal obtained by step 2, i.e., initially with transient state point
Shape computational methods impact response signal is pre-processed.Transient state calculation Method of Fractal is a kind of algorithm for being based on fractal dimension,
The method is processed time-domain signal using an instantaneous point shape measurer, can improve spectral purity；Specifically body process is：Assume light
The sampling interval of fine FBG impulse response signal is t, and the sample order of every group of response signal is followed successively by x1,x2,…,
xn, sequentially impulse response signal sample being once divided into into several subsegments according to this, each subsegment includes 128 samplings
Point；128 points of wherein k-th subsegment constitute sequence(j=1,2 ..., 128；K=1,2 ...), order
In above formula, NkWhen (Δ) represents that adopt width covers this signal segment for the grid of Δ, required grid number is fixed accordingly
Adopted instantaneously point shape amountFor：
Instantaneously divide shape amount using the parameter of fuzzy control as thisFunction ak, according to the feature of impact signal, can be with structure
Make some akWithFunctional relation is such as：
Using said method, can be with calculating corresponding mould according to arbitrarily one above-mentioned equation of selection during different functions section k
Paste control parameter ak, process being filtered to the signal of this function segment k with this parameter, most of stray wave can be eliminated to increase
Spectral purity.
Carry out WAVELET PACKET DECOMPOSITION again to the signal after transient state fractal calculation, the energy for extracting the 5th rank WAVELET PACKET DECOMPOSITION is total
With the characteristic information as positioning；Finally by the energy summation for extracting, K is designated as successively according to sensor numberA, KB, KC, KD,
KE, KF, KG；
Step 4：Optical fiber FBG shock response characteristic surface based on class two-dimensional normal distribution model is fitted
For some optical fiber FBG, after choosing the sample point impact on this sensor and adjacent sensors line
The response signal of gained, by these response signals, gained energy is sued for peace respectively Jing after the 5th rank WAVELET PACKET DECOMPOSITION；In conjunction with these
The position coordinates of sample shock point, sets up a shock characteristic comprising energy value information and coordinate information based on the sensor
Response sets：
{KI, p}
Wherein, i represents the sensor numbering A-G, and p represents sample point coordinate information；
According to optical fiber FBG sensitivity characteristic, fiber bragg grating center wavelength side-play amount with shock loading and sensor it
Between distance increase and exponentially attenuation trend；The shock loading response of sample point on two adjacent fiber FBG lines
The energy summation that signal Jing transient states fractal calculation and WAVELET PACKET DECOMPOSITION are extracted is presented attenuation trend with sample point distance increase, with one
The curve of dimension normal distribution mutually agrees with；
For represent certain optical fiber FBG its place monitoring subregion in characteristic equation, with reference to the sensor with
The attenuation curve of sample point transient load response characteristic on adjacent sensors line, can be configured to characterize optical fiber shock response
The class two-dimensional normal distribution surface model of characteristic；
The surface model of standard two-dimensional normal distribution is as follows：
Wherein σ1,σ2,μ1,μ2, ρ is constant, and it is σ to claim x, y to obey parameter1,σ2,μ1,μ2, the Two dimension normal distribution of ρ, often
This distribution is denoted as N (σ1,σ2,μ1,μ2,ρ)；
According to above-mentioned equation, first, standard two-dimensional normal distribution surface model is without otherness in x directions and y directions
, but the sensitivity characteristic of optical fiber FBG is being y directions and parallel to Fibre Optical Sensor axle perpendicular to Fibre Optical Sensor axial direction
There is notable difference to i.e. x directions；The susceptibility that the susceptibility in y directions is higher than on x directions is embodied in, equation is reflected in
Upper correlations of the variable y comprising information that be is higher than the information that variable x is included；Secondly, standard two-dimensional normal distribution is complete empty
Between on integration be equal to 1, and the model of required construction for optical fiber FBG sensitivity characteristic be fitted when then without the need for this limit
System；
Consider above-mentioned 2 factors, can build suitable for characterizing optical fiber FBG shock loading response characteristic
Class two-dimensional normal distribution model of fit；
The class two-dimensional normal distribution model of fit of optical fiber FBG shock loading in-Diane response feature is：
Wherein, x, y are sample impact point coordinates；z0, P, m1, m2, w1, w2It is Fitted parameter；
By shock characteristic response sets { KI, jIn element substitute into above-mentioned model of fit respectively and carry out class two-dimensional normal distribution
Surface fitting, tries to achieve above-mentioned Fitted parameter, can obtain fitting of each optical fiber FBG in respective subregion to be monitored
Toroidal function equation：
Fi(i=A～G)
Step 5：Shock loading location identification
If (5-1) any point is subject to transient impact, and the impact energy size and sample impact energy in monitored area
Amount is in the same size, and to the response signal of seven optical fiber FBGs transient state fractal calculation and WAVELET PACKET DECOMPOSITION the 5th are carried out successively
Rank energy and extraction, response K of difference Selection Center sensors AAResponse maximum in six optical fiber FBGs around
Value Ki.According to following Criterion of Selecting, it may be determined that the affiliated subregion in region to be measured that shock point loading position is located：
Work as KBFor during response characteristic signal maximum, then principium identification impact is carried in six optical fiber FBGs around
Lotus is located at I region；
Work as KCFor in six optical fiber FBGs around during response characteristic signal maximum, then principium identification shock point position
In II region；
Work as KDFor in six optical fiber FBGs around during response characteristic signal maximum, then principium identification shock point position
In III region；
Work as KEFor in six optical fiber FBGs around during response characteristic signal maximum, then principium identification shock point position
In IV region；
Work as KFFor in six optical fiber FBGs around during response characteristic signal maximum, then principium identification shock point position
In V region；
Work as KGFor in six optical fiber FBGs around during response characteristic signal maximum, then principium identification shock point position
In VI region；
(5-2) two optical fiber FBGs of monitoring subregion and its corresponding numbering, are can determine by (5-1), further according to
The corresponding shock response fitting surface function F of two optical fiber FBGs of correlationA, FiCarry out simultaneous solution；Concrete simultaneous mode
It is as follows：
Ⅰ：
Ⅱ：
Ⅲ：
Ⅳ：
Ⅴ：
Ⅵ：
Wherein, x, y are position coordinateses；KiIt is characteristic energy value；z0i, Pi, mi, ni, w1i, w2iObtained by being step 4 fitting
Class two-dimensional normal distribution fitting surface model of each optical fiber FBG in monitored area parameter, i is optical fiber FBG
Sensor number；I-VI sub- monitored area numbering to divide；
From fitting surface equation, x is tried to achieve, y there are six groups of solutions, if solution is real solution, keep constant, if solution
It is complex roots, then casts out imaginary part and retain its real part；Six groups of solutions are converted to after real number, according to the positive and negative values of x and y coordinates and
The subregion for being judged chooses suitable solution, and this solution is the position coordinates of shock loading.The selection rule of solution is specific as follows：
If the principium identification shock loading is located at I region, then it is minimum that of negative value, y-coordinate absolute value to choose x coordinate
Group solution shock loading position coordinates as determined by；
If the principium identification shock loading is located at II region, then choose x coordinate be negative value, y-coordinate be on the occasion of and the value
That larger group solution shock loading position coordinates as determined by；
If the principium identification shock loading is located at III region, then choose x coordinate be on the occasion of, y-coordinate be on the occasion of and the value
That larger group solution shock loading position coordinates as determined by；
If the principium identification shock loading is located at IV region, then it is minimum on the occasion of, y-coordinate absolute value to choose x coordinate
That group solution shock loading position coordinates as determined by；
If the principium identification shock loading is located at V region, then choose x coordinate to be on the occasion of, y-coordinate is negative value and definitely
That larger group solution shock loading position coordinates as determined by of value；
If the principium identification shock loading is located at VI region, then choose x coordinate be negative value, y-coordinate be negative value and definitely
That larger group solution shock loading position coordinates as determined by of value；
Advantages of the present invention：
First, in the Impact monitoring of actual large area region, if the region that each FBG can be allowed to detect is bigger,
Then the efficiency of FBG is higher.Monitoring efficiency is improved hence with FBG cellular layouts in large area structure Impact monitoring.In phase
In the case of with area of detection, compared to traditional square layout form, proposed honeycomb arrangement is used to reduce light
Fine FBG quantity.
Secondly, the feature extracting method based on transient state fractal calculation.Transient state fractal calculation is that one kind is directly carried out in time domain
Data calculate and filtering mode, it judges the size of transient component using the speed speed of signal intensity, signal intensity
Faster, transient component is higher, otherwise transient component is lower.So directly carry out process to transient component size to be equivalent to
Time domain conversion is carried out to signal.Optical fiber FBG impulse response signal can be effectively improved using transient state calculation Method of Fractal
Degree of purity, contributes to eliminating systematic error, so as to provide foundation for the accuracy of lifting feature signal extraction.
Again, the conventional structural impact load location method based on Fibre Optical Sensor, it usually needs set up in a large amount of priori
On the basis of knowledge.And the class two-dimensional normal distribution curved surface that this method employing can characterize FBG sensitivity characteristic is intended
Close, only need less sample point, you can the class two-dimensional normal distribution surface model obtained with fitting, to represent that related FBG is passed
The sensitivity characteristic of sensor, and shock loading position is obtained by solving simultaneous equation method.
Description of the drawings
Fig. 1 honeycomb fiber grating layouts and sample point are divided；
Fig. 2 Impact monitoring region divisions；
Fig. 3 class two-dimensional normal distribution surface model illustrations；
Fig. 4 shock loading positioning flows；
Specific embodiment
With reference to specific embodiment, further elucidate the present invention, it should be understood that these embodiments be only used for the present invention and not
For limiting the scope of the present invention, after reading this disclosure, those skilled in the art are to the various equivalent form of values of the invention
Modification falls within the application and invests claim and limits scope.
The shock loading positioning flow figure being fitted based on transient state fractal calculation and class two-dimensional normal distribution is as shown in figure 4, tool
The implementation steps of body are as follows：
Step one：Honeycomb monitoring network is arranged and corresponding monitored area divides
Monitored area is divided into into seven identical regular hexagon structural regions, and connects into honeycomb shape, i.e., middle one
Individual regular hexagon, six regular hexagons of surrounding；Each one optical fiber FBG of orthohexagonal center lateral arrangement, position
Sensors A is denoted as in middle Fibre Optical Sensor, remaining prolongs clockwise direction and is denoted as sensor B, C, D, E, F, G successively；
Middle regular hexagon is regarded as into six equilateral triangle areas, by each equilateral triangle, and positive six for being adjacent
Side shape is together as a monitored area, and whole monitored area is divided into six monitoring subregions, respectively monitor subregion I,
Ⅱ、Ⅲ、Ⅳ、Ⅴ、Ⅵ；
Wherein sensors A and sensor B monitor subregion I；Sensors A and sensor C monitoring subregions II；Sensors A
Subregion III is monitored with sensor D；Sensors A and sensor E monitoring subregions IV；Sensors A and sensor F monitoring subregions
Ⅴ；Sensors A and sensor G monitoring subregions VI；
Step 2：Impact sample point selection rule is gathered with impulse response signal
(2-1), a two-dimensional Cartesian coordinate system is set up on the impact region to be monitored of structural test piece, with fiber grating biography
Sensor A and E line direction as x-axis positive direction, using fiber-optic grating sensor A and B line direction as x-axis negative direction, with light
Fiber grating sensor A and sensor C, D midpoints line direction as y-axis positive direction, with fiber-optic grating sensor A and sensor F
With G midpoints line direction as y-axis negative direction；Impacting sample point selection rule is：On each two adjacent sensors line
It is even to take a little, and using these points as impact sample point；
(2-2) and then using jump bit these impact sample points are impacted successively using identical energy size；Respectively
Under record is per Secondary Shocks, the impulse response signal of distributing optical fiber sensing network；By the distributing optical fiber sensing under every Secondary Shocks
The impulse response signal of network is referred to as one group of impulse response signal；Every group of impulse response signal includes the corresponding punching of seven sensors
Hit the center wavelength shift amount of the sensor of response signal, i.e., seven；
Step 3：Shock response feature information extraction based on transient state fractal calculation
Successively feature information extraction is carried out to each group of impulse response signal obtained by step 2, i.e., initially with transient state point
Shape computational methods impact response signal is pre-processed.Transient state calculation Method of Fractal is a kind of algorithm for being based on fractal dimension,
The method is processed time-domain signal using an instantaneous point shape measurer, can improve spectral purity；Specifically body process is：Assume light
The sampling interval of fine FBG impulse response signal is t, and the sample order of every group of response signal is followed successively by x1,x2,…,
xn, sequentially impulse response signal sample being once divided into into several subsegments according to this, each subsegment includes 128 samplings
Point；128 points of wherein k-th subsegment constitute sequence(j=1,2 ..., 128；K=1,2 ...), order
In above formula, NkWhen (Δ) represents that adopt width covers this signal segment for the grid of Δ, required grid number is fixed accordingly
Adopted instantaneously point shape amountFor：
Instantaneously divide shape amount using the parameter of fuzzy control as thisFunction ak, according to the feature of impact signal, can be with structure
Make some akWithFunctional relation is such as：
Using said method, can be with calculating corresponding mould according to arbitrarily one above-mentioned equation of selection during different functions section k
Paste control parameter ak, process being filtered to the signal of this function segment k with this parameter, most of stray wave can be eliminated to increase
Spectral purity.
Carry out WAVELET PACKET DECOMPOSITION again to the signal after transient state fractal calculation, the energy for extracting the 5th rank WAVELET PACKET DECOMPOSITION is total
With the characteristic information as positioning；Finally by the energy summation for extracting, K is designated as successively according to sensor numberA, KB, KC, KD,
KE, KF, KG；
Step 4：Optical fiber FBG shock response characteristic surface based on class two-dimensional normal distribution model is fitted
For some optical fiber FBG, after choosing the sample point impact on this sensor and adjacent sensors line
The response signal of gained, by these response signals, gained energy is sued for peace respectively Jing after the 5th rank WAVELET PACKET DECOMPOSITION；In conjunction with these
The position coordinates of sample shock point, sets up a shock characteristic comprising energy value information and coordinate information based on the sensor
Response sets：
{KI, p}
Wherein, i represents the sensor numbering A-G, and p represents sample point coordinate information；
According to optical fiber FBG sensitivity characteristic, fiber bragg grating center wavelength side-play amount with shock loading and sensor it
Between distance increase and exponentially attenuation trend；The shock loading response of sample point on two adjacent fiber FBG lines
The energy summation that signal Jing transient states fractal calculation and WAVELET PACKET DECOMPOSITION are extracted is presented attenuation trend with sample point distance increase, with one
The curve of dimension normal distribution mutually agrees with；
For represent certain optical fiber FBG its place monitoring subregion in characteristic equation, with reference to the sensor with
The attenuation curve of sample point transient load response characteristic on adjacent sensors line, can be configured to characterize optical fiber shock response
The class two-dimensional normal distribution surface model of characteristic；
The surface model of standard two-dimensional normal distribution is as follows：
Wherein σ1,σ2,μ1,μ2, ρ is constant, and it is σ to claim x, y to obey parameter1,σ2,μ1,μ2, the Two dimension normal distribution of ρ, often
This distribution is denoted as N (σ1,σ2,μ1,μ2,ρ)；
According to above-mentioned equation, first, standard two-dimensional normal distribution surface model is without otherness in x directions and y directions
, but the sensitivity characteristic of optical fiber FBG is being y directions and parallel to Fibre Optical Sensor axle perpendicular to Fibre Optical Sensor axial direction
There is notable difference to i.e. x directions；The susceptibility that the susceptibility in y directions is higher than on x directions is embodied in, equation is reflected in
Upper correlations of the variable y comprising information that be is higher than the information that variable x is included, therefore by the exponential depth of y variables in equation
Number of times is changed to three times, to increase its variable information weight；Secondly, integration of the standard two-dimensional normal distribution in the total space is equal to
1, and the model of required construction is being used for when optical fiber FBG sensitivity characteristic is fitted then without the need for this restriction, accordingly, it would be desirable in side
Cheng Qian adds constant variables z0, for changing position of the model in reference axis according to the size of calculated energy value；
Consider above-mentioned 2 factors, can build suitable for characterizing optical fiber FBG shock loading response characteristic
Class two-dimensional normal distribution model of fit；
The class two-dimensional normal distribution model of fit of optical fiber FBG shock loading in-Diane response feature is：
The illustration of class two-dimensional normal distribution model is shown in Fig. 3, the value of each parameter is respectively in figure：z0=0, P=
1, m1=0, m2=0, w1=0.3, w2=0.3.
By shock characteristic response sets { KI, jIn element substitute into above-mentioned model of fit respectively and carry out class two-dimensional normal distribution
Surface fitting, tries to achieve above-mentioned Fitted parameter, can obtain fitting of each optical fiber FBG in respective subregion to be monitored
Toroidal function equation：
Fi(i=A～G)
Step 5：Shock loading location identification
If (5-1) any point is subject to transient impact, and the impact energy size and sample impact energy in monitored area
Amount is in the same size, and to the response signal of seven optical fiber FBGs transient state fractal calculation and WAVELET PACKET DECOMPOSITION the 5th are carried out successively
Rank energy and extraction, response K of difference Selection Center sensors AAResponse maximum in six optical fiber FBGs around
Value Ki.According to following Criterion of Selecting, it may be determined that the affiliated subregion in region to be measured that shock point loading position is located：
Work as KBFor during response characteristic signal maximum, then principium identification impact is carried in six optical fiber FBGs around
Lotus is located at I region；
Work as KCFor in six optical fiber FBGs around during response characteristic signal maximum, then principium identification shock point position
In II region；
Work as KDFor in six optical fiber FBGs around during response characteristic signal maximum, then principium identification shock point position
In III region；
Work as KEFor in six optical fiber FBGs around during response characteristic signal maximum, then principium identification shock point position
In IV region；
Work as KFFor in six optical fiber FBGs around during response characteristic signal maximum, then principium identification shock point position
In V region；
Work as KGFor in six optical fiber FBGs around during response characteristic signal maximum, then principium identification shock point position
In VI region；
(5-2) two optical fiber FBGs of monitoring subregion and its corresponding numbering, are can determine by (5-1), further according to
The corresponding shock response fitting surface function F of two optical fiber FBGs of correlationA, FiCarry out simultaneous solution；Concrete simultaneous mode
It is as follows：
Ⅰ：
Ⅱ：
Ⅲ：
Ⅳ：
Ⅴ：
Ⅵ：
Wherein, x, y are position coordinateses；KiIt is characteristic energy value；z0i, Pi, mi, ni, w1i, w2iObtained by being step 4 fitting
Class two-dimensional normal distribution fitting surface model of each optical fiber FBG in monitored area parameter, i is optical fiber FBG
Sensor number；I-VI sub- monitored area numbering to divide；
From fitting surface equation, x is tried to achieve, y there are six groups of solutions, if solution is real solution, keep constant, if solution
It is complex roots, then casts out imaginary part and retain its real part；Six groups of solutions are converted to after real number, according to the positive and negative values of x and y coordinates and
The subregion for being judged chooses suitable solution, and this solution is the position coordinates of shock loading.The selection rule of solution is specific as follows：
If the principium identification shock loading is located at I region, then it is minimum that of negative value, y-coordinate absolute value to choose x coordinate
Group solution shock loading position coordinates as determined by；
If the principium identification shock loading is located at II region, then choose x coordinate be negative value, y-coordinate be on the occasion of and the value
That larger group solution shock loading position coordinates as determined by；
If the principium identification shock loading is located at III region, then choose x coordinate be on the occasion of, y-coordinate be on the occasion of and the value
That larger group solution shock loading position coordinates as determined by；
If the principium identification shock loading is located at IV region, then it is minimum on the occasion of, y-coordinate absolute value to choose x coordinate
That group solution shock loading position coordinates as determined by；
If the principium identification shock loading is located at V region, then choose x coordinate to be on the occasion of, y-coordinate is negative value and definitely
That larger group solution shock loading position coordinates as determined by of value；
If the principium identification shock loading is located at VI region, then choose x coordinate be negative value, y-coordinate be negative value and definitely
That larger group solution shock loading position coordinates as determined by of value.
Claims (2)
1. a kind of Impact Location Method being fitted based on transient state fractal calculation and class two-dimensional normal distribution, it is characterised in that include
Following steps：
Step one：Honeycomb monitoring network is arranged and corresponding monitored area divides
Monitored area is divided into into seven identical regular hexagon structural regions, and connects into honeycomb shape, i.e., middle one just
Hexagon, six regular hexagons of surrounding；Each one optical fiber FBG of orthohexagonal center lateral arrangement, in being located at
Between Fibre Optical Sensor be denoted as sensors A, remaining prolongs and be denoted as sensor B, C, D, E, F, G successively clockwise；
Middle regular hexagon is regarded as into six equilateral triangle areas, by each equilateral triangle, and the regular hexagon being adjacent
Be together as a monitored area, whole monitored area is divided into six monitoring subregions, respectively monitor subregion I, II,
Ⅲ、Ⅳ、Ⅴ、Ⅵ；
Wherein sensors A and sensor B monitor subregion I；Sensors A and sensor C monitoring subregions II；Sensors A and biography
Sensor D monitors subregion III；Sensors A and sensor E monitoring subregions IV；Sensors A and sensor F monitoring subregions V；
Sensors A and sensor G monitoring subregions VI；
Step 2：Impact sample point selection rule is gathered with impulse response signal
(2-1), a two-dimensional Cartesian coordinate system is set up on the impact region to be monitored of structural test piece, with fiber-optic grating sensor
A and E lines direction as x-axis positive direction, using fiber-optic grating sensor A and B line direction as x-axis negative direction, with optical fiber light
Gate sensor A and sensor C, D midpoints line direction as y-axis positive direction, with fiber-optic grating sensor A and sensor F and G
Point line direction is used as y-axis negative direction；Impacting sample point selection rule is：Uniformly take on each two adjacent sensors line
Point, and using these points as impact sample point；
(2-2) and then using jump bit these impact sample points are impacted successively using identical energy size；Record respectively
Under per Secondary Shocks, the impulse response signal of distributing optical fiber sensing network；By the distributing optical fiber sensing network under every Secondary Shocks
Impulse response signal be referred to as one group of impulse response signal；Every group of impulse response signal includes that the respective impact of seven sensors is rung
The center wavelength shift amount of the sensor of induction signal, i.e., seven；
Step 3：Shock response feature information extraction based on transient state fractal calculation
Successively feature information extraction is carried out to each group of impulse response signal obtained by step 2, i.e., initially with transient state point shape meter
Calculation method impact response signal is pre-processed, and eliminates most of stray wave to increase spectral purity；
Carry out WAVELET PACKET DECOMPOSITION again to the signal after transient state fractal calculation, the energy summation for extracting the 5th rank WAVELET PACKET DECOMPOSITION is made
For the characteristic information of positioning；Finally by the energy summation for extracting, K is designated as successively according to sensor numberA, KB, KC, KD, KE,
KF, KG；
Step 4：Optical fiber FBG shock response characteristic surface based on class two-dimensional normal distribution model is fitted
For some optical fiber FBG, gained after the sample point impact on this sensor and adjacent sensors line is chosen
Response signal, by these response signals Jing after the 5th rank WAVELET PACKET DECOMPOSITION gained energy sue for peace respectively；In conjunction with these samples
The position coordinates of shock point, sets up a response of the shock characteristic comprising energy value information and coordinate information based on the sensor
Set：
{KI, p}
Wherein, i represents the sensor numbering A-G, and p represents sample point coordinate information；
According to optical fiber FBG sensitivity characteristic, fiber bragg grating center wavelength side-play amount is with shock loading and the spacing of sensor
From increase and exponentially attenuation trend；The shock loading response signal of sample point on two adjacent fiber FBG lines
The energy summation that Jing transient states fractal calculation and WAVELET PACKET DECOMPOSITION are extracted is presented attenuation trend with sample point distance increase, with it is one-dimensional just
The curve of state distribution mutually agrees with；
For represent certain optical fiber FBG its place monitoring subregion in characteristic equation, with reference to the sensor with it is adjacent
The attenuation curve of sample point shock loading response characteristic on sensor line, can be configured to characterize optical fiber shock response characteristic
Class two-dimensional normal distribution surface model；
The surface model of standard two-dimensional normal distribution is as follows：
Wherein σ1,σ2,μ1,μ2, ρ is constant, and it is σ to claim x, y to obey parameter1,σ2,μ1,μ2, the Two dimension normal distribution of ρ, often this
Individual distribution is denoted as N (σ1,σ2,μ1,μ2,ρ)；
According to above-mentioned equation, first, standard two-dimensional normal distribution surface model in x directions and y directions without otherness, but
The sensitivity characteristic of optical fiber FBG is being y directions and parallel to the axial i.e. x of Fibre Optical Sensor perpendicular to Fibre Optical Sensor axial direction
There is notable difference in direction；The susceptibility that the susceptibility in y directions is higher than on x directions is embodied in, is reflected on equation i.e.
The information that variable x is included is higher than for correlations of the variable y comprising information；Secondly, standard two-dimensional normal distribution is in the total space
Integration be equal to 1, and the model of required construction for optical fiber FBG sensitivity characteristic be fitted when then without the need for this limit；
Consider above-mentioned 2 factors, the class suitable for characterizing optical fiber FBG shock loading response characteristic can be built
Two dimension normal distribution model of fit；
The class two-dimensional normal distribution model of fit of optical fiber FBG shock loading in-Diane response feature is：
Wherein, x, y are sample impact point coordinates；z0, P, m1, m2, w1, w2It is Fitted parameter；
By shock characteristic response sets { KI, jIn element substitute into above-mentioned model of fit respectively and carry out class two-dimensional normal distribution curved surface
Fitting, tries to achieve above-mentioned Fitted parameter, can obtain fitting surface of each optical fiber FBG in respective subregion to be monitored
Functional equation：
Fi(i=A～G)
Step 5：Shock loading location identification
If (5-1) any point is subject to transient impact in monitored area, and the impact energy size is big with sample impact energy
It is little consistent, transient state fractal calculation and the rank energy of WAVELET PACKET DECOMPOSITION the 5th are carried out successively to the response signal of seven optical fiber FBGs
Amount and extraction, response K of difference Selection Center sensors AAAround response K maximum in six optical fiber FBGsi；
According to following Criterion of Selecting, it may be determined that the affiliated subregion of shock point loading position：
Work as KBFor during response characteristic signal maximum, then the principium identification shock loading is located in six optical fiber FBGs around
I region；
Work as KCFor during response characteristic signal maximum, then principium identification shock point is located at IIth area in six optical fiber FBGs around
Domain；
Work as KDFor during response characteristic signal maximum, then principium identification shock point is located at IIIth area in six optical fiber FBGs around
Domain；
Work as KEFor during response characteristic signal maximum, then principium identification shock point is located at IVth area in six optical fiber FBGs around
Domain；
Work as KFFor during response characteristic signal maximum, then principium identification shock point is located at Vth area in six optical fiber FBGs around
Domain；
Work as KGFor during response characteristic signal maximum, then principium identification shock point is located at VIth area in six optical fiber FBGs around
Domain；
(5-2) two optical fiber FBGs of monitoring subregion and its corresponding numbering, are can determine by (5-1), further according to correlation
The corresponding shock response fitting surface function F of two optical fiber FBGsA, FiCarry out simultaneous solution；Concrete simultaneous mode is as follows：
Ⅰ：
Ⅱ：
Ⅲ：
Ⅳ：
Ⅴ：
Ⅵ：
Wherein, x, y are position coordinateses；KiIt is characteristic energy value；z0i, Pi, mi, ni, w1i, w2iIt is every obtained by step 4 fitting
The parameter of class two-dimensional normal distribution fitting surface model of the individual optical fiber FBG in monitored area, i is optical fiber FBG sensings
Device is numbered；I-VI sub- monitored area numbering to divide；
From fitting surface equation, x is tried to achieve, y there are six groups of solutions, if solution is real solution, keep constant, if Xie Shifu
Number solution, then cast out imaginary part and retain its real part；Six groups of solutions are converted to after real number, according to the positive and negative values of x and y coordinates and are sentenced
Fixed subregion chooses suitable solution, and this solution is the position coordinates of shock loading；The selection rule of solution is specific as follows：
If the principium identification shock loading is positioned at I region, then chooses x coordinate and solve for that minimum group of negative value, y-coordinate absolute value
The shock loading position coordinates as determined by；
If the principium identification shock loading is located at II region, then choose x coordinate be negative value, y-coordinate be on the occasion of and the value is larger
That group solution as determined by shock loading position coordinates；
If the principium identification shock loading is located at III region, then choose x coordinate be on the occasion of, y-coordinate be on the occasion of and the value is larger
That group solution as determined by shock loading position coordinates；
If the principium identification shock loading is located at IV region, then it is on the occasion of that minimum group of, y-coordinate absolute value to choose x coordinate
Solution shock loading position coordinates as determined by；
If the principium identification shock loading is located at V region, then choose x coordinate be on the occasion of, y-coordinate be negative value and absolute value compared with
That big group solution shock loading position coordinates as determined by；
If the principium identification shock loading is located at VI region, then choose x coordinate be negative value, y-coordinate be negative value and absolute value compared with
That big group solution shock loading position coordinates as determined by.
2. the Impact Location Method being fitted based on transient state fractal calculation and class two-dimensional normal distribution according to claim 1,
It is characterized in that：Described transient state calculation Method of Fractal is that the method is using instantaneously point shape based on a kind of algorithm of fractal dimension
Measurer is processed optical fiber FBG shock response time-domain signal, can improve spectral purity；Specifically body process is：It is false
If the sampling interval of optical fiber FBG impulse response signal is t, the sample order of every group of response signal is followed successively by x1,
x2,…,xn, sequentially impulse response signal sample being once divided into into several subsegments according to this, each subsegment includes 128
Sampled point；128 points of wherein k-th subsegment constitute sequenceOrder
In above formula, NkWhen (Δ) represents that adopt width covers this signal segment for the grid of Δ, required grid number defines accordingly wink
Time-division shape amountFor：
Instantaneously divide shape amount using the parameter of fuzzy control as thisFunction ak, according to the feature of impact signal, if can construct
Dry akWithFunctional relation is such as：
Using said method, can be with calculating corresponding Fuzzy Control according to arbitrarily one above-mentioned equation of selection during different functions section k
Parameter a processedk, process being filtered to the signal of this function segment k with this parameter, most of stray wave can be eliminated to increase signal
Purity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610902301.2A CN106643484B (en) | 2016-10-17 | 2016-10-17 | The Impact Location Method being fitted based on transient state fractal calculation and class two-dimensional normal distribution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610902301.2A CN106643484B (en) | 2016-10-17 | 2016-10-17 | The Impact Location Method being fitted based on transient state fractal calculation and class two-dimensional normal distribution |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106643484A true CN106643484A (en) | 2017-05-10 |
| CN106643484B CN106643484B (en) | 2018-12-07 |
Family
ID=58856869
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610902301.2A Active CN106643484B (en) | 2016-10-17 | 2016-10-17 | The Impact Location Method being fitted based on transient state fractal calculation and class two-dimensional normal distribution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106643484B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108195326A (en) * | 2017-12-14 | 2018-06-22 | 大连理工大学 | A kind of shock localization method based on empirical mode decomposition |
| CN110220452A (en) * | 2018-03-01 | 2019-09-10 | 中兴通讯股份有限公司 | A kind of localization method and system |
| CN113049410A (en) * | 2021-02-04 | 2021-06-29 | 南京航空航天大学 | Composite material laminate optical fiber impact position identification method based on strain nonlinear weighting |
| CN113049409A (en) * | 2021-02-04 | 2021-06-29 | 南京航空航天大学 | Thin-wall flexible arm structure optical fiber impact object identification method based on signal frequency band energy |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5798521A (en) * | 1996-02-27 | 1998-08-25 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Apparatus and method for measuring strain in bragg gratings |
| US6492636B1 (en) * | 1997-08-19 | 2002-12-10 | The University Of Maryland | Large scale high speed multiplexed optical fiber sensor network |
| CN104019739A (en) * | 2014-05-08 | 2014-09-03 | 南京航空航天大学 | Plate structure impact load positioning method based on distributed fiber bragg grating sensing network |
| CN104776966A (en) * | 2015-04-01 | 2015-07-15 | 南京航空航天大学 | Plate structure impact monitoring method based on fractal theory |
-
2016
- 2016-10-17 CN CN201610902301.2A patent/CN106643484B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5798521A (en) * | 1996-02-27 | 1998-08-25 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Apparatus and method for measuring strain in bragg gratings |
| US6492636B1 (en) * | 1997-08-19 | 2002-12-10 | The University Of Maryland | Large scale high speed multiplexed optical fiber sensor network |
| CN104019739A (en) * | 2014-05-08 | 2014-09-03 | 南京航空航天大学 | Plate structure impact load positioning method based on distributed fiber bragg grating sensing network |
| CN104776966A (en) * | 2015-04-01 | 2015-07-15 | 南京航空航天大学 | Plate structure impact monitoring method based on fractal theory |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108195326A (en) * | 2017-12-14 | 2018-06-22 | 大连理工大学 | A kind of shock localization method based on empirical mode decomposition |
| CN108195326B (en) * | 2017-12-14 | 2019-08-20 | 大连理工大学 | A kind of shock localization method based on empirical mode decomposition |
| CN110220452A (en) * | 2018-03-01 | 2019-09-10 | 中兴通讯股份有限公司 | A kind of localization method and system |
| CN113049410A (en) * | 2021-02-04 | 2021-06-29 | 南京航空航天大学 | Composite material laminate optical fiber impact position identification method based on strain nonlinear weighting |
| CN113049409A (en) * | 2021-02-04 | 2021-06-29 | 南京航空航天大学 | Thin-wall flexible arm structure optical fiber impact object identification method based on signal frequency band energy |
| CN113049409B (en) * | 2021-02-04 | 2022-04-22 | 南京航空航天大学 | Thin-wall flexible arm structure optical fiber impact object identification method based on signal frequency band energy |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106643484B (en) | 2018-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106643484A (en) | Impact positioning method based on transient fractal calculation and similar two-dimensional normal distribution fitting | |
| CN104019739B (en) | Plate structure impact load location method based on distributed fiber grating sensing network | |
| CN104062677B (en) | Multifunctional comprehensive integrated high-precision intelligent micro-seismic monitoring system | |
| CN111761409A (en) | Multi-sensor numerical control machine tool cutter wear monitoring method based on deep learning | |
| CN104614179B (en) | A kind of gearbox of wind turbine state monitoring method | |
| CN101556724A (en) | Safety management system of optical fiber perimeter and pattern recognition method thereof | |
| CN103954368B (en) | A kind of narrow band light demodulating system based on photoelectronic detecting array and demodulation method thereof | |
| CN107885940A (en) | A kind of signal characteristic extracting methods for distributed optical fiber vibration sensing system | |
| CN104776966A (en) | Plate structure impact monitoring method based on fractal theory | |
| CN108399696A (en) | Intrusion behavior recognition methods and device | |
| CN105844632A (en) | Rice plant identification and positioning method based on machine visual sense | |
| CN108008151A (en) | A kind of moving state identification method and system based on 3-axis acceleration sensor | |
| CN106875630A (en) | A kind of wearable fall detection method and system based on hierarchical classification | |
| CN108061635A (en) | Harden structure Impact monitoring method based on Teager energy operators and Sample Entropy | |
| CN111986238A (en) | Concrete arch dam modal shape identification method based on unmanned aerial vehicle video shooting | |
| CN108051126B (en) | A kind of Varying-thickness Composite Laminated Plate under Low-Velocity Impact Thin interbed system and its working method | |
| CN101067659B (en) | Remote sensing image sorting method | |
| CN109541037B (en) | Loader structure main crack screening method based on compromise and competition mechanism | |
| CN101788664A (en) | Land degradation response unit-oriented high-spectrum land degradation information extracting method | |
| Gagnon et al. | Zooplankton sample variability in a tidal estuary: An interpretative model 1 | |
| CN106546319B (en) | The clamped plate Impact monitoring method calculated based on fiber grating transient response difference | |
| CN115526424A (en) | Machine tool spindle Z-direction thermal deformation prediction method based on multi-source heterogeneous information fusion | |
| CN106446503A (en) | Method for identifying time-varying working mode of auto-covariance matrix recursive principal component analysis with forgetting factor | |
| CN115564338A (en) | Multi-working-condition-oriented numerical control machine tool key component cluster performance evaluation method | |
| CN201740627U (en) | Electric measurement device for dynamic balance |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |