CN110261052A - Using power hammer excitation and photogrammetric Modal Analysis of Structures system and method - Google Patents
Using power hammer excitation and photogrammetric Modal Analysis of Structures system and method Download PDFInfo
- Publication number
- CN110261052A CN110261052A CN201910532534.1A CN201910532534A CN110261052A CN 110261052 A CN110261052 A CN 110261052A CN 201910532534 A CN201910532534 A CN 201910532534A CN 110261052 A CN110261052 A CN 110261052A
- Authority
- CN
- China
- Prior art keywords
- point
- power hammer
- image
- power
- modal
- 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
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M7/00—Vibration-testing of structures; Shock-testing of structures
Abstract
The present invention relates to a kind of using power hammer excitation and photogrammetric Modal Analysis of Structures system and method, using power hammer excitation and the photogrammetric test of contactless Produced by Modal and analysis system, it is characterized in that, overcome the previous analysis being only capable of mostly based on photogrammetric modal analysis system to structure progress OMA, the synchronous acquisition that the present invention passes through power hammer and video, it can reach the measurement of whole audience displacement synchronous, it can accurately estimate the structural frequency response Jacobian matrix hammered into shape by power and video determines, give full play to the advantage of EMA, improve the precision and reliability of Modal Parameter Identification.Realize the Non-contact modal test and analysis to the quick high accuracy of structure.
Description
Technical field
The invention belongs to Produced by Modal testing field, it is related to a kind of shaking using power hammer excitation with photogrammetric structure
Dynamic modal analysis system and method.
Background technique
Experimental modal analysis is the basic hand for obtaining the inherent characteristics such as engineering structure vibration frequency, damping and Mode Shape
Section.In practical engineering applications, according to the difference of data acquisition, the method for being usually used in Modal Analysis of Structures is broadly divided into two
Class: first is that the traditional modal analysis based on input and output, i.e. experimental modal analysis (EMA)；Second is that being based only on the mode point of output
Analysis, i.e. operational modal analysis (OMA).In the past 20 years, operational modal analysis increasingly becomes research hotspot, due to operation mode point
Analysis is based only on output response, that is, assumes that input is white noise, current all modal identification methods based on output all exist
It developed under this hypothesis.And input encountered in engineering reality is likely to be unsatisfactory for this it is assumed that making to identify that work is general
All over encountering some puzzlements, such as mode determine that rank is difficult, false mode excessively, identification error.Input signal is obtained using power hammer
It can effectively solve the problems, such as this.
Be commonly used to acquisition output has acceleration transducer, laser vibration measurer and high-speed camera.Acceleration transducer is
The most frequently used is also most popular mould measurement tool, has the advantages such as direct measurement, measurement accuracy height.But using adding
Velocity sensor measurement can introduce additional mass, should not use contact type measurement tool under some special test conditions.Compared to
Acceleration transducer, laser vibration measurer are a kind of non-contacting measurement devices, and measurement accuracy is high, and measurement frequency band is broad.It is insufficient
Place is, takes a long time the measurement data for obtaining scanning area, time-consuming and laborious.Using high-speed camera can overcome with
The deficiency of upper two kinds of measuring tools, this measurement method are also referred to as photogrammetric technology.
It is existing based on photogrammetric modal analysis system, mostly can only independent analysis video or image, can only be to knot
Structure carries out OMA, cannot analyze the input signal of video and power hammer simultaneously, can not play the advantage of EMA.The present invention is existing to overcome
Defect, discloses a kind of using power hammer excitation and photogrammetric Modal Analysis of Structures system, can analyze power hammer simultaneously
The output of input and high speed camera provides reliable contactless experimental modal analysis techniques for engineering structure.
Summary of the invention
Technical problems to be solved
In order to avoid the shortcomings of the prior art, the present invention propose it is a kind of using power hammer excitation and photogrammetric structure
Vibration Modal Analysis System and method.
Technical solution
It is a kind of using power hammer excitation and photogrammetric Modal Analysis of Structures system, it is characterised in that including power hammer,
IEPE signal conditioner, data collector, trigger circuit and high-speed camera；IEPE type pressure sensing is equipped in the power hammer
Device；The output end that power hammers signal into shape connects IEPE signal conditioner, the output end of IEPE signal conditioner draw all the way signal to touching
The input terminal of Power Generation Road, another way connect data processing PC unit by data collector；The output end of trigger circuit connects high
The output of the triggering route of fast video camera, high-speed camera connects data processing PC unit；When power hammer applies measured structure
When exciting force, IEPE type pressure sensor converts electric signal for force signal and exports, and the electric signal of output passes through IEPE signal tune
Reason device and data collector are input to data processing PC unit；Meanwhile the output of IEPE signal conditioner is opened by trigger circuit
Dynamic high-speed camera, high-speed camera real time shooting power hammer applies image data when exciting force to measured structure, and transmits
To data processing PC unit；Two kinds of signals are handled and carry out Produced by Modal to measured structure by data processing PC unit
Analysis.
First stage amplifier is connected between IEPE signal conditioner and data collector.
It is a kind of using described using power hammer excitation and photogrammetric Modal Analysis of Structures system obtains vibration displacement
Method, it is characterised in that steps are as follows:
Step 1, the relative resolution mu-factor for measuring visual field: it is with the circular markers of testee by high-speed camera
Visual field is measured, acquires the data image for containing circular mark with high-speed camera, its circle is found out using circle detection algorithm
Outline position and diameter pixel p, then the relative resolution coefficient of the measurement point bedrFor the straight of circular mark
Diameter, the diameter of a circle that p is obtained with circle detection algorithm, unit are pixels；
Step 2: the sample rate that power hammer excitation is arranged is 3200hz, sampling time 4s；The acquisition frame of high-speed camera is set
Rate is 3200fps, and the acquisition frame number after triggering is 12800；Power hammer taps tested circular markers, and power hammer excitation obtains power letter
Number, and trigger high-speed camera acquisition module and obtain image of video data；
Step 3, the detection of image of video data characteristic point: using first frame image as reference frame, in reference frame image
Mark point is area-of-interest Region of Interest, ROI, is carried out using Harris Corner Detection Algorithm to ROI region
Detection, obtains the position coordinates of several pixels；Using the position coordinates of pixel as the starting point of tracking；
Step 4 carries out target following to characteristic point: target following uses Kanade-Lucas-Tomasi algorithm, when from t
Quarter, acquired image sequence are S=(It,It+1,...,It+k), u=(x, y) indicates the position of u point on t moment image
Coordinate is denoted as It(x,y)
Motion excursion amount d=(d occurs in t to t+1 momentx,dy), then 1 point of the u position on t+1 moment image is remembered
For It+1(x+dx,y+dy)；In the neighborhood w centered on point u, there are difference functions:
ε (d)=ε (dx,dy)=∫w(It(x,y)-It+1(x+dx,y+dy))2dw
Purpose is to calculate motion excursion amount d, so that the value of ε (d) is minimum；This process makes ε (d) using Newton iteration method
Value convergence, if beyond iteration setting number do not restrain still, then it is assumed that the position point tracking failure；
Step 5, screening target following track: image ItAs the start image of final track, tracks and be positive since u point
To tracking, until image It+k, then motion profile is (ut,ut+1,...,ut+k), it is denoted asIt+k→It
It is denoted as backward tracingBut starting tracking point isDistal point is ut；
It enablesThen forward and reverse two kinds of tracing path error are as follows:
Given threshold is δ=0.001, unit: pixel；
WhenWhen, it is believed that the track of tracking is effective, otherwise it is assumed that tracking is failed；
The vibration displacement of step 6, measurement point: according to the relative resolution coefficient μ of step 1 measurement point, step 4 is accorded with
Close the pursuit path T of required precisionk, then the vibration displacement of the measurement point are as follows: x=μ Tk。
In the step 3, Harris Corner Detection Algorithm is replaced using SURF and minimal characteristic detection algorithm, is obtained several
The position coordinates of a pixel.
In the step 4, the convergent the number of iterations of value of ε (d) is made to be set as 30~50 using Newton iteration method.
It is a kind of using described using power hammer excitation and photogrammetric Modal Analysis of Structures system and the acquisition
Vibration displacement carry out model analysis method, it is characterised in that steps are as follows:
Step (1), frequency response function calculate: being provided with n measurement point on being measured structure, be fixed on o-th of measurement point
Locate (o≤n), exciting k times, each exciting force is denoted as fi(i=1,2 ... k)；Every exciting is primary, measures all measurements of structure
The vibration displacement of point are as follows: X={ x1,x2,...xn, after exciting k times, the vibration displacement of each measurement point are as follows: Xi(i=1,
...2k)；
Frequency response function isWherein,Indicate XiWith fiCross-power spectrum estimation,Indicate fiFrom
Power Spectral Estimation；
Frequency response function matrix is
It is right respectivelyFull phase time shift phase difference correction is carried out, new frequency response function is obtained
Step (2), Modal Parameter Identification: to k times obtained frequency response function matrixIt is averaged, i.e.,
By HoIt is write asWherein No(ω) is molecule multinomial, and D (ω) is denominator polynomials；
ThenWherein, N is multinomial order, Zj(ω)
For basic function, matrix coefficient AjAnd BojIt is exactly the parameter finally to be estimated；
Determine denominator coefficients matrix α (α={ A1,A2,...An}T), solve the eigenvalue λ of the adjoint matrix of αrWith feature to
V is measured, pole is exactly eigenvalue λr, mode participation vector LrIt is exactly the last line of V matrix；The natural frequency ω of r rankrWith
DampingratioζrIt is calculated by following formula:
The Mode Shape φ of r rankrIt is calculated by following formula:
Wherein, [LR] is lower discrepance, and [UR] is upper discrepance.
Beneficial effect
It is proposed by the present invention a kind of using power hammer excitation and photogrammetric Modal Analysis of Structures system and method, it adopts
With power hammer excitation and the test of photogrammetric contactless Produced by Modal and analysis system, it is characterized in that, overcome with
Past is only capable of the analysis to structure progress OMA based on photogrammetric modal analysis system mostly, and the present invention passes through power and hammers into shape and regard
The synchronous acquisition of frequency can reach the measurement of whole audience displacement synchronous, can accurately estimate the structural frequency response function hammered into shape by power and video determines
Matrix gives full play to the advantage of EMA, improves the precision and reliability of Modal Parameter Identification.Realize the quick high accuracy to structure
Non-contact modal test and analysis.
Detailed description of the invention
Fig. 1 is Hardware Design.
Fig. 2 is the mould measurement schematic diagram of cantilever beam.
Fig. 3 is the test chart of relative resolution.
Fig. 4 is the flow chart that vibration displacement calculates.
Fig. 5 is the difference calculation process runs of target tracking algorism.
Fig. 6 is model analysis flow chart.
Fig. 7 is the steady state picture calculated result of cantilever beam structure mode of oscillation
Specific embodiment
Now in conjunction with embodiment, attached drawing, the invention will be further described:
Hardware design:
1, power hammer excitation module: including power hammer, IEPE signal conditioner, charge amplifier, data acquisition equipment and data
Acquisition software.The effect of power hammer excitation system is to apply exciting force to measured structure and acquire excitation force signal.The system institute
The power hammer used, its working principle is that being integrated with an IEPE type pressure sensor inside it, the tup of power hammer taps tested
Structure can trigger pressure sensor work, convert electric signal output for force signal, the electric signal of output is by data acquisition equipment
It collects in experiment computer, then has been converted to force signal divided by the sensitivity that power is hammered into shape.It should be noted that IEPE type passes
Sensor cannot generally be directly connected to data acquisition equipment, need a constant current source power supply, i.e. IEPE signal conditioner.Therefore,
In the present system, the output end of power hammer signal first accesses IEPE signal conditioner, then draws output end from IEPE signal conditioner
To computer.When the lesser model of selection sensitivity, the conversion ratio that force signal is converted into electric signal is low, generates electric signal
Amplitude is lower.At this point, needing to add a charge amplifier for the ease of doing subsequent force signal analysis.
2, camera acquisition module is triggered, including IEPE signal conditioner, trigger circuit, camera external trigger line, is taken the photograph at a high speed
Camera and high-speed video acquisition software.The effect of triggering camera acquisition system is to guarantee that modal analysis system can collect simultaneously
Pumping signal and response signal trigger high speed camera synchronous acquisition when power, which is hammered into shape, taps measured structure.In power hammer excitation system
In, the signal of power hammer passes through IEPE signal conditioner, and it is electric to triggering to draw signal all the way from the output end of IEPE signal conditioner
The input terminal on road is integrated with voltage comparator and charge amplifier in trigger circuit, the reference voltage with voltage comparator setting
It is compared, when input terminal voltage is greater than reference voltage, voltage comparator module is started to work, and exports a Transistor-Transistor Logic level；It is no
Then, voltage comparator module does not work.Wherein, voltage comparator module uses TLV3501 chip, and delay time can achieve
4.5ns, it is ensured that being up in sample frequency is not in trigger delay within 1MHz.The amplitude of output voltage is equal to triggering electricity
The supply voltage on road, usual supply voltage are 3.5~5V.The output end of trigger circuit is connected to camera external trigger route,
The external trigger mouth for triggering another termination high speed camera of route, when triggering mouth receives Transistor-Transistor Logic level, high-speed camera starts
Acquire video.
Method and step:
1, vibration displacement calculates:
(1) relative resolution for measuring visual field calculates: the relative resolution α for measuring visual field is indicated are as follows:In order to avoid pattern distortion caused by camera lens as far as possible, only in the neck of measurement point
Domain calculates it first to resolution ratio.Specific steps: it is placed around and its circular mark in the same plane in point to be measured, it is known that
The diameter of circular mark is dr=15mm finds out its circular contour position and diameter pixel p using circle detection algorithm, then surveys
Relative resolution coefficient at amount point is represented by
(2) characteristic point detects: after collecting high-speed video source, extracting the first frame image of video as reference frame.With
For some mark point in reference frame image, mark point region (ROI region) is manually selected, is calculated with Harris Corner Detection
Method detects ROI region, obtains the position coordinates of several pixels.Other than Harris Corner Detection Algorithm, also
SURF and minimal characteristic detection algorithm can be used.Using the position coordinates of these pixels as the starting point of tracking.
(3) carry out target following to characteristic point: target following uses Kanade-Lucas-Tomasi (KLT) algorithm: assuming that
The position of reference point is (x, y), and motion excursion amount d=(ξ, η) has occurred, then the position of the second frame is (x+ ξ, y+ η).KLT is calculated
Method assumes that the brightness at two o'clock is consistent, i.e. I (x, y, 0)=I (x+ ξ, y+ η, t).Enable J (X)=I (x, y, 0), I (X+d)=
I (x+ ξ, y+ η, t), then
J (X)=I (X+d)+n (X) (1)
Wherein n (X) is noise.The difference value of brightness can indicate at two o'clock are as follows:
ε=∫w[I(X+d)-J(X)]2wdX (2)
Wherein, w is the size of the domain window of setting；ε is denoted as residual error, is the letter about the quadratic power of movement warp d
Number.
Firstly, doing first order Taylor expansion to I (X+d), obtain:
I (X+d)=I (X)+gd (3)
WhereinSubstitution formula (2) can obtain:
ε=∫w[I(X)+g·d-J(X)]2WdX=∫w(h+g·d)2wdX (4)
Wherein h=I (X)-J (X).Keep residual error minimum, last for enabling (4) formula is zero to the first differential of d, it may be assumed that
∫w(h+gd) (5) gwdA=0
Due to (gd) g=(ggT) d, and think that d is continuously, then in w:
(∫wggTWdA) d=- ∫whdgwdA (6)
It can simplify as Gd=e, wherein G=∫wggTWdA, e=∫w(J-I) gwdA, e are to calculate residual error, solve d and use
Newton iteration method.Experiments verify that setting the number of iterations as 30 times.When result convergence, it is believed that obtained solution is accurate solution；If
It does not restrain, then it is assumed that position point tracking failure.
(4) it screens target following track: when carrying out characteristic point detection to each measurement point region, many textures can be obtained
Characteristic point.However, the tracking accuracy of not each characteristic point is very high, or even pixel can be lost when tracking certain characteristic points
Location information.Carrying out screening hypothesis image sequence therefore, it is necessary to the pursuit path to these characteristic points is S=(It,
It+1,...,It+k), wherein u=(x1,y1) indicate t moment picture point at position.Image ItStarting figure as final track
Picture is tracked since u point, until image It+k, then motion profile is (ut,ut+1,...,ut+k).This process is referred to as positive tracking,
It is denoted asEqually, It+k→ItIt is denoted as backward tracingBut starting tracking point isDistal point is ut.It enablesThen forward and reverse two kinds of tracing path error are as follows:Setting
Threshold value is δ=0.001 (unit: pixel), whenWhen, it is believed that the track of tracking is effectively, otherwise it is assumed that chasing after
Track failure.
(5) estimate the vibration displacement of measurement point: the relative resolution factor alpha of measurement point, step are obtained by step (1)
(4) the pursuit path T for meeting required precision is obtainedk, then the vibration displacement of the measurement point may be expressed as: x=α Tk。
2, experimental modal analysis (EMA).Experimental modal analysis is broadly divided into two steps: first is that resultant force hammer input and
The output of camera calculates the frequency response function of structure；Second is that estimating the modal parameter of structure, i.e. frequency, damping by frequency response function
And the vibration shape.
(1) frequency response function calculates: if being provided with n measurement point on being measured structure, being fixed on o-th of measurement point (o
≤ n), exciting k times, each exciting force is denoted as fi(i=1,2 ... k).Every exciting is primary, measures all measurement points of structure
Vibration displacement are as follows: X={ x1,x2,...xn, after exciting k times, the vibration displacement of each measurement point are as follows: Xi(i=1,2...k).Frequently
Ringing function can be by XiIt is indicated with the ratio between the Fourier transformation of F, i.e.,In experimental modal analysis, H is estimatedi(f)
Method generally use H1The estimation technique, i.e.,Wherein,Indicate XiWith fiCross-power spectrum estimation,Table
Show fiAuto-power spectrum estimation.Then a complete column for frequency response function matrix are represented byWithIt is right respectively for referenceCarry out full phase time shift phase
Potential difference correction, obtains new frequency response function
(2) Modal Parameter Identification: to k times obtained frequency response function matrixIt is averaged, i.e.,By HoIt is write asWherein No(ω) is molecule multinomial, and D (ω) is that denominator is more
Item formula.No(ω) and D (ω) can be expressed asWherein, N
For multinomial order, Zj(ω) is basic function, matrix coefficient AjAnd BojIt is exactly the parameter finally to be estimated.Denominator coefficients have been determined
Matrix α (α={ A1,A2,...An}T), the eigenvalue λ of the adjoint matrix by solving αrWith feature vector V, pole is exactly special
Value indicative λr, mode participation vector LrIt is exactly the last line of V matrix.
The natural frequency ω of r rankrAnd dampingratioζrIt can be acquired by formula (7), the Mode Shape φ of r rankrIt can be by formula (8)
It acquires:
Wherein, [LR] is lower discrepance, and [UR] is upper discrepance.
By taking the freedom of cantilever beam-Free Modal test as an example.
Using power hammer excitation and photogrammetric Produced by Modal detecting and analysing system, including Hardware Design (figure
And software system analysis 1).The freedom of cantilever beam-Free Modal test schematic diagram is as shown in Figure 2.Wherein: cantilever beam 1, power hammer 2,
Signal conditioner 3, trigger circuit 4, data acquisition equipment 5, high-speed camera 6, light-supplementing system 7, Modal Analysis of Structures meter
Calculation machine 8.
Steps are as follows for specific experiment:
1. experimental setup.The both ends of cantilever beam spring rope is hung, free-free test mode of simulation.Test side
Surface mount mark point to adjust camera position as measurement point, make to measure the mark point side that visual field concentrates on cantilever beam,
As shown in Figure 1.The acquisition frame rate of camera is 3200fps, acquisition time 4s；
2. calculating the relative resolution of measurement visual field.In the top of cantilever beam, a specific circular mark is placed, it is known that
The diameter of circular mark is 15mm.An image is acquired with high speed camera, as shown in Figure 3.Image is obtained using loop truss algorithm
Pixel number shared by middle circular mark, is calculated relative resolution.It is influenced caused by pattern distortion when to reduce measurement, every
A measurement point all carries out this operation, obtains the relative resolution of each measurement point.
3. power hammers signal and vision signal synchronous acquisition into shape.Synchronous acquisition needs the hardware used effectively to hammer into shape, high-speed camera,
Camera external trigger line, data acquisition equipment, IEPE signal conditioner, trigger circuit and experiment computer.To guarantee acquisition
The consistency of data, the sample rate of setting power hammer excitation system are 3200hz, sampling time 4s, setting triggering camera acquisition system
Acquisition frame rate be 3200fps, acquisition frame number after triggering is 12800.When power hammer taps third measurement point, power hammer excitation
Module obtains force signal, and triggering camera acquisition module obtains video source.
4. vibration displacement calculates.Step 3 is obtained into video source inputted vibration displacement computing system, obtains all measurement points
Coordinate position versus time curve.The specific flow chart of video algorithm processing system is as shown in Figure 4.It misses target following track
The calculating process of difference is as shown in Figure 5.
Modal parameter calculates.Force signal and vibration displacement signal after comprehensive repeatedly excitation, by experimental modal analysis system
Obtain the modal parameter of structure.Model analysis process is as shown in Figure 6.Fig. 7 is the steady state picture of cantilever beam structure mode of oscillation, table 1
It is the frequency resultant of acceleration transducer and high speed camera identification.Table 2 is the MAC of acceleration transducer and high speed camera measurement
Value.
The frequency estimation result of 1 acceleration transducer of table and high speed camera
| Order | Acceleration transducer | High speed camera |
| 1 | 60.834 | 60.875 |
| 2 | 330.344 | 330.524 |
| 3 | 544.699 | 545.965 |
| 4 | 812.488 | 814.126 |
| 5 | 1134.711 | 1138.104 |
| 7 | 1511.746 | 1514.374 |
The MAC value calculated result of 2 acceleration transducer of table and high speed camera
| Order | 1 | 2 | 3 | 4 | 5 | 6 |
| 1 | 98.07 | 1.52 | 2.26 | 1.67 | 0.55 | 1.72 |
| 2 | 3.99 | 97.96 | 1.72 | 2.63 | 2.11 | 2.84 |
| 3 | 2.41 | 2.06 | 97.54 | 1.57 | 2.93 | 2.41 |
| 4 | 1.79 | 1.74 | 2.93 | 93.82 | 2.64 | 0.91 |
| 5 | 0.37 | 1.32 | 3.04 | 2.43 | 90.08 | 3.02 |
| 6 | 1.98 | 1.32 | 1.86 | 1.78 | 3.48 | 86.52 |
Claims (6)
1. a kind of using power hammer excitation and photogrammetric Modal Analysis of Structures system, it is characterised in that including power hammer,
IEPE signal conditioner, data collector, trigger circuit and high-speed camera；IEPE type pressure sensing is equipped in the power hammer
Device；The output end that power hammers signal into shape connects IEPE signal conditioner, the output end of IEPE signal conditioner draw all the way signal to touching
The input terminal of Power Generation Road, another way connect data processing PC unit by data collector；The output end of trigger circuit connects high
The output of the triggering route of fast video camera, high-speed camera connects data processing PC unit；When power hammer applies measured structure
When exciting force, IEPE type pressure sensor converts electric signal for force signal and exports, and the electric signal of output passes through IEPE signal tune
Reason device and data collector are input to data processing PC unit；Meanwhile the output of IEPE signal conditioner is opened by trigger circuit
Dynamic high-speed camera, high-speed camera real time shooting power hammer applies image data when exciting force to measured structure, and transmits
To data processing PC unit；Two kinds of signals are handled and carry out Produced by Modal to measured structure by data processing PC unit
Analysis.
2. being existed according to claim 1 using power hammer excitation and photogrammetric Modal Analysis of Structures system, feature
In: between IEPE signal conditioner and data collector connect first stage amplifier.
3. a kind of obtained using power hammer excitation with photogrammetric Modal Analysis of Structures system using as claimed in claim 1 or 2
The method for obtaining vibration displacement, it is characterised in that steps are as follows:
Step 1, measure visual field relative resolution mu-factor: by high-speed camera with the circular markers of testee be measurement
Visual field acquires the data image for containing circular mark with high-speed camera, finds out its circular contour using circle detection algorithm
Position and diameter pixel p, then the relative resolution coefficient of the measurement point bedrFor the diameter of circular mark, p is used
The diameter of a circle that circle detection algorithm obtains, unit are pixels；
Step 2: the sample rate that power hammer excitation is arranged is 3200hz, sampling time 4s；The acquisition frame rate that high-speed camera is arranged is
3200fps, the acquisition frame number after triggering are 12800；Power hammer taps tested circular markers, and power hammer excitation obtains force signal,
And it triggers high-speed camera acquisition module and obtains image of video data；
Step 3, the detection of image of video data characteristic point: using first frame image as reference frame, with the label in reference frame image
Point is area-of-interest Region of Interest, ROI, is detected using Harris Corner Detection Algorithm to ROI region,
Obtain the position coordinates of several pixels；Using the position coordinates of pixel as the starting point of tracking；
Step 4 carries out target following to characteristic point: target following uses Kanade-Lucas-Tomasi algorithm, opens from t moment
Begin, acquired image sequence is S=(It,It+1,...,It+k), u=(x, y) indicates that the position of u point on t moment image is sat
Mark, is denoted as It(x,y)
Motion excursion amount d=(d occurs in t to t+1 momentx,dy), then 1 point of the u position on t+1 moment image is denoted as It+1
(x+dx,y+dy)；In the neighborhood w centered on point u, there are difference functions:
ε (d)=ε (dx,dy)=∫w(It(x,y)-It+1(x+dx,y+dy))2dw
Purpose is to calculate motion excursion amount d, so that the value of ε (d) is minimum；This process makes the value of ε (d) using Newton iteration method
Convergence, if not restrained still beyond iteration setting number, then it is assumed that position point tracking failure；
Step 5, screening target following track: image ItAs the start image of final track, tracks since u point and chased after for forward direction
Track, until image It+k, then motion profile is (ut,ut+1,...,ut+k), it is denoted asIt+k→ItIt is denoted as
Backward tracingBut starting tracking point isDistal point is ut；
It enablesThen forward and reverse two kinds of tracing path error are as follows:
Given threshold is δ=0.001, unit: pixel；
WhenWhen, it is believed that the track of tracking is effective, otherwise it is assumed that tracking is failed；
The vibration displacement of step 6, measurement point: according to the relative resolution coefficient μ of step 1 measurement point, step 4 obtains meeting essence
Spend desired pursuit path Tk, then the vibration displacement of the measurement point are as follows: x=μ Tk。
4. according to the method described in claim 3, it is characterized by: being calculated in the step 3 using SURF and minimal characteristic detection
Method replaces Harris Corner Detection Algorithm, obtains the position coordinates of several pixels.
5. according to the method described in claim 3, it is characterized by: making ε's (d) using Newton iteration method in the step 4
It is worth convergent the number of iterations and is set as 30~50.
6. it is a kind of using as claimed in claim 1 or 2 using power hammer excitation and photogrammetric Modal Analysis of Structures system, with
And the method that the vibration displacement of the acquisition of claim 3 or 4 carries out model analysis, it is characterised in that steps are as follows:
Step (1), frequency response function calculate: being provided with n measurement point on being measured structure, be fixed on o-th of measurement point (o
≤ n), exciting k times, each exciting force is denoted as fi(i=1,2 ... k)；Every exciting is primary, measures all measurement points of structure
Vibration displacement are as follows: X={ x1,x2,...xn, after exciting k times, the vibration displacement of each measurement point are as follows: Xi(i=1,2...k)；
Frequency response function isWherein,Indicate XiWith fiCross-power spectrum estimation,Indicate fiFrom power
Power estimation；
Frequency response function matrix is
It is right respectivelyFull phase time shift phase difference correction is carried out, new frequency response function is obtained
Step (2), Modal Parameter Identification: to k times obtained frequency response function matrixIt is averaged, i.e.,
Ho is write asWherein No (ω) is molecule multinomial, and D (ω) is denominator polynomials；
ThenWherein, N is multinomial order, Zj(ω) is base
Function, matrix coefficient AjAnd BojIt is exactly the parameter finally to be estimated；
Determine denominator coefficients matrix α (α={ A1,A2,...An}T), solve the eigenvalue λ of the adjoint matrix of αrWith feature vector V,
Its pole is exactly eigenvalue λr, mode participation vector LrIt is exactly the last line of V matrix；The natural frequency ω of r rankrAnd damping
Compare ζrIt is calculated by following formula:
The Mode Shape φ of r rankrIt is calculated by following formula:
Wherein, [LR] is lower discrepance, and [UR] is upper discrepance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910532534.1A CN110261052B (en) | 2019-06-19 | 2019-06-19 | Structural vibration mode analysis system and method adopting force hammer excitation and photogrammetry |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910532534.1A CN110261052B (en) | 2019-06-19 | 2019-06-19 | Structural vibration mode analysis system and method adopting force hammer excitation and photogrammetry |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110261052A true CN110261052A (en) | 2019-09-20 |
| CN110261052B CN110261052B (en) | 2020-11-06 |
Family
ID=67919484
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910532534.1A Active CN110261052B (en) | 2019-06-19 | 2019-06-19 | Structural vibration mode analysis system and method adopting force hammer excitation and photogrammetry |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110261052B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111189598A (en) * | 2019-12-31 | 2020-05-22 | 中汽研汽车检验中心（天津）有限公司 | Free modal test method for car wheel |
| CN112444365A (en) * | 2020-11-30 | 2021-03-05 | 哈尔滨工业大学 | Satellite solar wing substrate unfolding low-frequency modal testing method based on force hammer swing method and laser Doppler method |
| CN114184271A (en) * | 2021-11-30 | 2022-03-15 | 西安航天动力研究所 | High-frequency vibration mode identification method and system based on motion amplification and digital image correlation |
| CN114739350A (en) * | 2022-04-12 | 2022-07-12 | 招商局公路信息技术（重庆）有限公司 | Pavement dynamic tire pressure detector calibration method and system based on modal excitation |
| CN114993451A (en) * | 2022-06-23 | 2022-09-02 | 无锡中鼎集成技术有限公司 | Low-frequency vibration testing system and measuring method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013163175A1 (en) * | 2012-04-23 | 2013-10-31 | Rassini Frenos, S.A. De C.V. | Method and apparatus for measuring damping in a workpiece |
| CN103954221A (en) * | 2014-05-08 | 2014-07-30 | 哈尔滨工业大学 | Binocular photogrammetry method of large flexible structure vibration displacement |
| CN105403143A (en) * | 2015-10-28 | 2016-03-16 | 苏州西博三维科技有限公司 | Measurement method and system thereof for acquiring displacement changes of earthquake simulator |
| CN106124034A (en) * | 2016-09-07 | 2016-11-16 | 湖南科技大学 | Thin-wall part operation mode based on machine vision test device and method of testing |
| US20180283847A1 (en) * | 2017-03-29 | 2018-10-04 | Shimadzu Corporation | Vibration measurement device |
| CN109060284A (en) * | 2018-08-07 | 2018-12-21 | 广东工业大学 | A kind of Experimental modal analysis method based on DIC technology |
-
2019
- 2019-06-19 CN CN201910532534.1A patent/CN110261052B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013163175A1 (en) * | 2012-04-23 | 2013-10-31 | Rassini Frenos, S.A. De C.V. | Method and apparatus for measuring damping in a workpiece |
| CN103954221A (en) * | 2014-05-08 | 2014-07-30 | 哈尔滨工业大学 | Binocular photogrammetry method of large flexible structure vibration displacement |
| CN105403143A (en) * | 2015-10-28 | 2016-03-16 | 苏州西博三维科技有限公司 | Measurement method and system thereof for acquiring displacement changes of earthquake simulator |
| CN106124034A (en) * | 2016-09-07 | 2016-11-16 | 湖南科技大学 | Thin-wall part operation mode based on machine vision test device and method of testing |
| US20180283847A1 (en) * | 2017-03-29 | 2018-10-04 | Shimadzu Corporation | Vibration measurement device |
| CN109060284A (en) * | 2018-08-07 | 2018-12-21 | 广东工业大学 | A kind of Experimental modal analysis method based on DIC technology |
Non-Patent Citations (1)
| Title |
|---|
| 戴俊: "车用制动零件模态检测系统设计", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111189598A (en) * | 2019-12-31 | 2020-05-22 | 中汽研汽车检验中心（天津）有限公司 | Free modal test method for car wheel |
| CN111189598B (en) * | 2019-12-31 | 2022-06-14 | 中汽研汽车检验中心（天津）有限公司 | Free modal test method for car wheel |
| CN112444365A (en) * | 2020-11-30 | 2021-03-05 | 哈尔滨工业大学 | Satellite solar wing substrate unfolding low-frequency modal testing method based on force hammer swing method and laser Doppler method |
| CN112444365B (en) * | 2020-11-30 | 2023-08-29 | 哈尔滨工业大学 | Satellite solar wing substrate unfolding low-frequency mode testing method |
| CN114184271A (en) * | 2021-11-30 | 2022-03-15 | 西安航天动力研究所 | High-frequency vibration mode identification method and system based on motion amplification and digital image correlation |
| CN114184271B (en) * | 2021-11-30 | 2023-06-27 | 西安航天动力研究所 | High-frequency vibration mode identification method and system based on motion amplification and digital image correlation |
| CN114739350A (en) * | 2022-04-12 | 2022-07-12 | 招商局公路信息技术（重庆）有限公司 | Pavement dynamic tire pressure detector calibration method and system based on modal excitation |
| CN114739350B (en) * | 2022-04-12 | 2023-09-26 | 招商局公路信息技术（重庆）有限公司 | Method and system for calibrating road surface dynamic tire pressure detector based on modal excitation |
| CN114993451A (en) * | 2022-06-23 | 2022-09-02 | 无锡中鼎集成技术有限公司 | Low-frequency vibration testing system and measuring method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110261052B (en) | 2020-11-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110261052A (en) | Using power hammer excitation and photogrammetric Modal Analysis of Structures system and method | |
| US9927222B2 (en) | Position/orientation measurement apparatus, measurement processing method thereof, and non-transitory computer-readable storage medium | |
| CN104981105B (en) | A kind of quickly accurate detection and method for correcting error for obtaining element central and deflection angle | |
| JP5384316B2 (en) | Displacement measuring device, displacement measuring method, and displacement measuring program | |
| CN110675455B (en) | Natural scene-based self-calibration method and system for vehicle body looking-around camera | |
| AU2020101196A4 (en) | Method and system for testing working modality of thin-walled member based on monocular visual optical flow tracking | |
| CN109341549A (en) | A kind of camera system displacement monitoring improved method based on data fusion | |
| CN107357426A (en) | A kind of motion sensing control method for virtual reality device | |
| Lv et al. | A point tracking method of TDDM for vibration measurement and large-scale rotational motion tracking | |
| CN107945166B (en) | Binocular vision-based method for measuring three-dimensional vibration track of object to be measured | |
| CN101794395B (en) | Image matching positioning method based on Sift algorithm | |
| Zhu et al. | Full-field modal identification using reliability-guided frequency-domain-based digital image correlation method based on multi-camera system | |
| JP2016081121A (en) | Information processing unit, information processing method and program | |
| CN110532725B (en) | Engineering structure mechanical parameter identification method and system based on digital image | |
| JP2010009236A (en) | Plane area estimation device and program | |
| CN114266835A (en) | Deformation monitoring control method and system for non-measuring camera | |
| JP5230354B2 (en) | POSITIONING DEVICE AND CHANGED BUILDING DETECTION DEVICE | |
| CN114442808A (en) | Pose tracking module testing method, device, equipment, system and medium | |
| JP4611708B2 (en) | High-precision mark position / posture detection device | |
| Torres et al. | A practical algorithm to correct geometrical distortion of image acquisition cameras | |
| CN116008970B (en) | Method for verifying radar null value inversion accuracy based on video image | |
| JP5359477B2 (en) | Road area estimation apparatus and program | |
| US20220138471A1 (en) | Load estimatation apparatus and method | |
| CN115638731B (en) | Super-resolution-based vibrating table test computer vision displacement measurement method | |
| CN116309712B (en) | Functional motion analysis system and method based on depth camera |
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 |