CN105527016B - A kind of power transformer Separation by vibration method and system based on time-frequency than distribution - Google Patents
A kind of power transformer Separation by vibration method and system based on time-frequency than distribution Download PDFInfo
- Publication number
- CN105527016B CN105527016B CN201511030951.4A CN201511030951A CN105527016B CN 105527016 B CN105527016 B CN 105527016B CN 201511030951 A CN201511030951 A CN 201511030951A CN 105527016 B CN105527016 B CN 105527016B
- Authority
- CN
- China
- Prior art keywords
- time
- signal
- vibration
- frequency
- group
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000000926 separation method Methods 0.000 title claims abstract description 29
- 239000011159 matrix material Substances 0.000 claims abstract description 54
- 238000012216 screening Methods 0.000 claims abstract description 49
- 230000009466 transformation Effects 0.000 claims abstract description 17
- 238000004364 calculation method Methods 0.000 claims description 29
- 238000005315 distribution function Methods 0.000 claims description 22
- 230000008569 process Effects 0.000 claims description 11
- 238000010276 construction Methods 0.000 claims description 5
- 238000002156 mixing Methods 0.000 claims description 4
- 238000003805 vibration mixing Methods 0.000 claims 1
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 25
- 238000004804 winding Methods 0.000 description 21
- 239000003921 oil Substances 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 4
- 230000005284 excitation Effects 0.000 description 3
- 229910052742 iron Inorganic materials 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 229910000976 Electrical steel Inorganic materials 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
This application discloses a kind of power transformer Separation by vibration method and system based on time-frequency than distribution, this method includes：Using preset mixed signal screening conditions, first group of mixed signal and second group of mixed signal are filtered out from vibration mixed signal；Calculate above-mentioned when frequency ratio between the two；The amplitude best estimate and phase best estimate of frequency ratio when calculating separately；Using amplitude best estimate and phase best estimate, hybrid matrix estimated value is constructed, and invert to it, obtain corresponding inverse matrix；Fourier transform is carried out to vibration mixed signal, the result accordingly obtained is multiplied by above-mentioned inverse matrix, the Fourier transformation of basket vibration source signal and core vibration source signal is obtained as a result, and to result progress inversefouriertransform, being correspondingly made available the time-domain signal of basket vibration and core vibration.The application realizes the purpose for separating basket vibration signal and core vibration signal from tank surface vibration signal.
Description
Technical field
The present invention relates to signal processing technology field, more particularly to a kind of power transformer based on time-frequency than distribution vibrates
Separation method and system.
Background technology
Large-scale power transformer is one of key equipment particularly significant and valuable in network system, and safe operation is to protecting
Demonstrate,prove power grid security, reliable most important.In order to avoid causing the major accident of transformer, Ren Menxu due to mechanical structure defect
Analyzing processing is carried out to the vibration in the power transformer course of work.
Winding and iron core are the critical pieces of transformer, while being also the major failure component of transformer, winding and iron core
Or generate the main source block of transformer vibration.At runtime, the magnetostriction of silicon steel sheet generates the vibration of iron core to transformer,
And the electromagnetic force of load current then makes winding vibrate.Winding and the vibration of iron core connect firmware and absolutely by inside transformer
Edge oil is transmitted to oil tank of transformer surface, is superimposed in tank surface and generates vibration.
The vibration analysis means of power transformer are usually only rested in the detection to tank surface vibration signal at present,
And the vibration state of tank surface vibration signal reflecting voltage device entirety, and fail to embody the vibration state and iron of winding respectively
The vibration state of core.Since the principle and state feature of winding and core vibration are different, in order to effectively monitor winding
With the state and failure of iron core, it is necessary to detach basket vibration signal and core vibration signal from tank surface vibration signal
It comes.
In summary as can be seen that how by basket vibration signal and core vibration signal from tank surface vibration signal
It separates, to realize that effective monitor to the respective working condition of winding and iron core is current urgent problem to be solved.
Invention content
In view of this, the purpose of the present invention is to provide a kind of power transformer Separation by vibration sides based on time-frequency than distribution
Method and system have reached the mesh for separating basket vibration signal and core vibration signal from tank surface vibration signal
, to realize effective monitoring to the respective working condition of winding and iron core.Its concrete scheme is as follows：
A kind of power transformer Separation by vibration method based on time-frequency than distribution, including：
The vibration signal of the tank surface of power transformer is acquired, vibration mixed signal is obtained；
Using preset mixed signal screening conditions, filtered out from the vibration mixed signal first group of mixed signal and
Second group of mixed signal；
Calculate the when frequency ratio between first group of mixed signal and second group of mixed signal；
When calculating separately described the amplitude best estimate of frequency ratio and it is described when frequency ratio phase best estimate；
Using the amplitude best estimate and the phase best estimate, hybrid matrix estimated value is constructed, and to institute
It states hybrid matrix estimated value to invert, obtains corresponding inverse matrix；
Fourier transform is carried out to vibration mixed signal, the result accordingly obtained is multiplied by the inverse matrix, obtains winding
The Fourier transformation of source signal and core vibration source signal is vibrated as a result, and by the basket vibration source signal and core vibration source
The Fourier transform results of signal carry out inversefouriertransform, are correspondingly made available the time-domain signal and core vibration of basket vibration
Time-domain signal；
Wherein, the expression formula of the hybrid matrix estimated value is：
Wherein, b21For with basket vibration source signal s1(n) corresponding amplitude best estimate, b22To believe with core vibration source
Number s2(n) corresponding amplitude best estimate, μ21For with the basket vibration source signal s1(n) corresponding phase best estimate
Value, μ22For with the core vibration source signal s2(n) corresponding phase best estimate.
Preferably, the mixed signal screening conditions are：Voltage effective value in first group of mixed signal with it is described
The absolute value of the difference of voltage effective value in second group of mixed signal is less than preset voltage value, and first group of mixed signal
In current effective value and second group of mixed signal in the absolute value of the difference of current effective value be more than pre-set current value.
Preferably, it is described calculate between first group of mixed signal and second group of mixed signal when frequency ratio mistake
Journey, including：
By when frequency ratio calculation formula, calculate between first group of mixed signal and second group of mixed signal when
Frequency ratio；Wherein, frequency ratio calculation formula is when described：
Wherein, V1(n, ω) is the Short Time Fourier Transform of first group of mixed signal；V2(n, ω) is described second group
The short time discrete Fourier transform of mixed signal；S1(n, ω) is the basket vibration source signal s1(n) Short Time Fourier Transform；S2
(n, ω) is the core vibration source signal s2(n) Short Time Fourier Transform.
Preferably, the process of the amplitude best estimate of frequency ratio when calculating described, including：
Using Dan Yuan time-frequencies area prescreening condition, pre- menu source time-frequency area is filtered out from the vibration mixed signal；Its
In, Dan Yuan time-frequencies area prescreening condition is：| | β (n, ω) |-| β (n+1, ω) | | < θ1；Wherein, | β (n, ω) | it indicates
When frequency ratio amplitude；θ1For pre-set first threshold；
Clustering is carried out to the when frequency ratio amplitude in the pre- menu source time-frequency area, is built for characterizing amplitude distribution journey
The distribution function of degree；Wherein, the distribution function is：
The local maximum for calculating the distribution function obtains the first maximum and the second maximum；
The tangent value for calculating separately first maximum and second maximum, be correspondingly made available the first tangent value and
Second tangent value, the amplitude best estimate b of frequency ratio when first tangent value is determined as21, second tangent value is true
The amplitude best estimate b of frequency ratio when being set to22；
Wherein, feature vector
Position vector
Preferably, the process of the phase best estimate of frequency ratio when calculating described, including：
Using the first Zu Danyuan time-frequencies area screening conditions, filters out from the pre- menu source time-frequency area and shake with the winding
Dynamic source signal s1(n) corresponding first Zu Danyuan time-frequencies area SSW1；
Using the second Zu Danyuan time-frequencies area screening conditions, filters out from the pre- menu source time-frequency area and shake with the iron core
Dynamic source signal s2(n) corresponding second Zu Danyuan time-frequencies area SSW2；
Calculate separately the first Zu Danyuan time-frequencies area SSW1With the second Zu Danyuan time-frequencies area SSW2Phase mean value,
It is correspondingly made available first phase mean value and second phase mean value, the phase of frequency ratio is best when the first phase mean value is determined as
Estimated value μ21, the phase best estimate μ of frequency ratio when the second phase mean value is determined as22；
Wherein, the first Zu Danyuan time-frequencies area screening conditions are：||β(n,ω)|-b21| < θ2, second group of list source
Time-frequency area screening conditions are：||β(n,ω)|-b22| < θ2, θ2For pre-set second threshold；
The calculation formula of the first phase mean value is：Wherein,It is described first
Zu Danyuan time-frequencies area SSW1When frequency ratio phase value, N1For the first Zu Danyuan time-frequencies area SSW1The number in Zhong Danyuan time-frequencies area；
The calculation formula of the second phase mean value is：Wherein,It is described second
Zu Danyuan time-frequencies area SSW2When frequency ratio phase value, N2For the second Zu Danyuan time-frequencies area SSW2The number in Zhong Danyuan time-frequencies area.
The invention also discloses a kind of power transformer Separation by vibration systems based on time-frequency than distribution, including：
Signal pickup assembly, the vibration signal for the tank surface to power transformer are acquired, and it is mixed to obtain vibration
Close signal；
Signal screening module is screened for utilizing preset mixed signal screening conditions from the vibration mixed signal
Go out first group of mixed signal and second group of mixed signal；
When frequency ratio computing module, for calculate between first group of mixed signal and second group of mixed signal when
Frequency ratio；
Best estimate computing module, when for calculating separately described the amplitude best estimate of frequency ratio and it is described when frequency ratio
Phase best estimate；
Matrix construction module, for utilizing the amplitude best estimate and the phase best estimate, construction mixing
Matrix Estimation value, and invert to the hybrid matrix estimated value, obtain corresponding inverse matrix；
The result accordingly obtained is multiplied by institute by signal separation module for carrying out Fourier transform to vibration mixed signal
Inverse matrix is stated, obtains the Fourier transformation of basket vibration source signal and core vibration source signal as a result, and by the basket vibration
The Fourier transform results of source signal and core vibration source signal carry out inversefouriertransform, be correspondingly made available basket vibration when
The time-domain signal of domain signal and core vibration；
Wherein, the expression formula of the hybrid matrix estimated value is：
Wherein, b21For with basket vibration source signal s1(n) corresponding amplitude best estimate, b22To believe with core vibration source
Number s2(n) corresponding amplitude best estimate, μ21For with the basket vibration source signal s1(n) corresponding phase best estimate
Value, μ22For with the core vibration source signal s2(n) corresponding phase best estimate.
Preferably, the mixed signal screening conditions are：Voltage effective value in first group of mixed signal with it is described
The absolute value of the difference of voltage effective value in second group of mixed signal is less than preset voltage value, and first group of mixed signal
In current effective value and second group of mixed signal in the absolute value of the difference of current effective value be more than pre-set current value.
Preferably, frequency ratio computing module when described, frequency ratio calculation formula when being specifically used for passing through calculate described first group and mix
Close the when frequency ratio between signal and second group of mixed signal；Wherein, frequency ratio calculation formula is when described：
Wherein, V1(n, ω) is the Short Time Fourier Transform of first group of mixed signal；V2(n, ω) is described second group
The short time discrete Fourier transform of mixed signal；S1(n, ω) is the basket vibration source signal s1(n) Short Time Fourier Transform；S2
(n, ω) is the core vibration source signal s2(n) Short Time Fourier Transform.
Preferably, the best estimate computing module includes amplitude best estimate computing module and phase best estimate
It is worth computing module；Wherein, the amplitude best estimate computing module includes：
Dan Yuan time-frequencies area prescreening unit, for utilizing Dan Yuan time-frequencies area prescreening condition, from the vibration mixed signal
In filter out pre- menu source time-frequency area；Wherein, Dan Yuan time-frequencies area prescreening condition is：||β(n,ω)|-|β(n+1,ω)|
| < θ1；Wherein, | β (n, ω) | the amplitude of frequency ratio when expression；θ1For pre-set first threshold；
Distribution function structural unit, for carrying out clustering to the when frequency ratio amplitude in the pre- menu source time-frequency area,
Build the distribution function for characterizing amplitude distribution degree；Wherein, the distribution function is：
Local maximum computing unit, the local maximum for calculating the distribution function, obtain the first maximum and
Second maximum；
Tangent value computing unit, the tangent value for calculating separately first maximum and second maximum, phase
Obtain the first tangent value and the second tangent value with answering, the amplitude best estimate of frequency ratio when first tangent value is determined as
b21, the amplitude best estimate b of frequency ratio when second tangent value is determined as22；
Wherein, feature vector
Position vector
Preferably, the phase best estimate computing module includes：
First Zu Danyuan time-frequencies area screening unit, for utilizing the first Zu Danyuan time-frequencies area screening conditions, from the pre-selection
It is filtered out in Dan Yuan time-frequencies area and the basket vibration source signal s1(n) corresponding first Zu Danyuan time-frequencies area SSW1；
Second Zu Danyuan time-frequencies area screening unit, for utilizing the second Zu Danyuan time-frequencies area screening conditions, from the pre-selection
It is filtered out in Dan Yuan time-frequencies area and the core vibration source signal s2(n) corresponding second Zu Danyuan time-frequencies area SSW2；
Phase average calculation unit, for calculating separately the first Zu Danyuan time-frequencies area SSW1With second group of list source
Time-frequency area SSW2Phase mean value, first phase mean value and second phase mean value are correspondingly made available, by the first phase mean value
The phase best estimate μ of frequency ratio when being determined as21, the phase best estimate of frequency ratio when the second phase mean value is determined as
μ22；
Wherein, the first Zu Danyuan time-frequencies area screening conditions are：||β(n,ω)|-b21| < θ2, second group of list source
Time-frequency area screening conditions are：||β(n,ω)|-b22| < θ2, θ2For pre-set second threshold；
The calculation formula of the first phase mean value is：Wherein,It is described first
Zu Danyuan time-frequencies area SSW1When frequency ratio phase value, N1For the first Zu Danyuan time-frequencies area SSW1The number in Zhong Danyuan time-frequencies area；
The calculation formula of the second phase mean value is：Wherein,It is described second
Zu Danyuan time-frequencies area SSW2When frequency ratio phase value, N2For the second Zu Danyuan time-frequencies area SSW2The number in Zhong Danyuan time-frequencies area.
In the present invention, power transformer Separation by vibration method, including：To the vibration signal of the tank surface of power transformer
It is acquired, obtains vibration mixed signal；Using preset mixed signal screening conditions, the is filtered out from vibration mixed signal
One group of mixed signal and second group of mixed signal；Calculate the when frequency ratio between first group of mixed signal and second group of mixed signal；
When calculating separately the amplitude best estimate of frequency ratio and when frequency ratio phase best estimate；Utilize amplitude best estimate and phase
Position best estimate, constructs hybrid matrix estimated value, and invert to hybrid matrix estimated value, obtains corresponding inverse matrix；
Fourier transform is carried out to vibration mixed signal, the result accordingly obtained is multiplied by above-mentioned inverse matrix, obtains basket vibration source letter
Number and core vibration source signal Fourier transformation as a result, and by the Fourier of basket vibration source signal and core vibration source signal
Transformation results carry out inversefouriertransform, are correspondingly made available the time-domain signal of basket vibration and the time-domain signal of core vibration.It can
See, the present invention just calculates first group after filtering out first group of mixed signal and second group of mixed signal in vibration mixed signal
When frequency ratio between mixed signal and second group of mixed signal, amplitude best estimate when by calculating this corresponding to frequency ratio and
Phase best estimate, and then corresponding hybrid matrix estimated value is built, the hybrid matrix estimated value is then utilized, winding is shaken
The time-domain signal of dynamic time-domain signal and core vibration is separated from vibration mixed signal, has thus been reached basket vibration
The purpose that signal and core vibration signal are separated from tank surface vibration signal, it is each to winding and iron core to realize
From effective monitoring of working condition.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of power transformer Separation by vibration method flow based on time-frequency than distribution disclosed by the embodiments of the present invention
Figure；
Fig. 2 is a kind of power transformer Separation by vibration system structure based on time-frequency than distribution disclosed by the embodiments of the present invention
Schematic diagram；
Fig. 3 is a kind of specifically power transformer Separation by vibration system based on time-frequency than distribution disclosed by the embodiments of the present invention
System structural schematic diagram.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a kind of power transformer Separation by vibration methods based on time-frequency than distribution, referring to Fig. 1
Shown, this method includes：
Step S11：The vibration signal of the tank surface of power transformer is acquired, vibration mixed signal is obtained；
Step S12：Using preset mixed signal screening conditions, first group of mixing letter is filtered out from vibration mixed signal
Number and second group of mixed signal；
Step S13：Calculate the when frequency ratio between first group of mixed signal and second group of mixed signal；
Step S14：When calculating separately the amplitude best estimate of frequency ratio and when frequency ratio phase best estimate；
Step S15：Using amplitude best estimate and phase best estimate, hybrid matrix estimated value is constructed, and to mixed
It closes Matrix Estimation value to invert, obtains corresponding inverse matrix；
Step S16：Fourier transform is carried out to vibration mixed signal, the result accordingly obtained is multiplied by above-mentioned inverse matrix,
The Fourier transformation of basket vibration source signal and core vibration source signal is obtained as a result, and basket vibration source signal and iron core shake
The Fourier transform results of dynamic source signal carry out inversefouriertransform, and the time-domain signal and iron core for being correspondingly made available basket vibration shake
Dynamic time-domain signal.
Wherein, the expression formula of hybrid matrix estimated value is：
Wherein, b21For with basket vibration source signal s1(n) corresponding amplitude best estimate, b22To believe with core vibration source
Number s2(n) corresponding amplitude best estimate, μ21For with basket vibration source signal s1(n) corresponding phase best estimate, μ22
For with core vibration source signal s2(n) corresponding phase best estimate.
In the present embodiment, mixed signal screening conditions are：Voltage effective value in first group of mixed signal and second group it is mixed
The absolute value of the difference for closing the voltage effective value in signal is less than preset voltage value, and the electric current in first group of mixed signal is effective
Value and the absolute value of the difference of the current effective value in second group of mixed signal are more than pre-set current value.Wherein, above-mentioned predeterminated voltage
The occurrence of value needs finally so that the voltage effective value in first group of mixed signal has with the voltage in second group of mixed signal
Valid value is close or identical；The occurrence of above-mentioned pre-set current value needs finally to make the current effective value in first group of mixed signal
Absolute value of the difference between the current effective value in second group of mixed signal is sufficiently large, namely is more than specific current thresholds.
In the embodiment of the present invention, power transformer Separation by vibration method, including：To shaking for the tank surface of power transformer
Dynamic signal is acquired, and obtains vibration mixed signal；Using preset mixed signal screening conditions, sieved from vibration mixed signal
Select first group of mixed signal and second group of mixed signal；Calculate between first group of mixed signal and second group of mixed signal when
Frequency ratio；When calculating separately the amplitude best estimate of frequency ratio and when frequency ratio phase best estimate；Utilize amplitude best estimate
Value and phase best estimate construct hybrid matrix estimated value, and invert to hybrid matrix estimated value, obtain corresponding inverse
Matrix；Fourier transform is carried out to vibration mixed signal, the result accordingly obtained is multiplied by inverse matrix, obtains basket vibration source letter
Number and core vibration source signal Fourier transformation as a result, and by the Fourier of basket vibration source signal and core vibration source signal
Transformation results carry out inversefouriertransform, are correspondingly made available the time-domain signal of basket vibration and the time-domain signal of core vibration.
As it can be seen that the embodiment of the present invention filters out first group of mixed signal and second group of mixed signal from vibration mixed signal
Afterwards, the when frequency ratio between first group of mixed signal and second group of mixed signal is just calculated, when by calculating this corresponding to frequency ratio
Amplitude best estimate and phase best estimate, and then corresponding hybrid matrix estimated value is built, then utilize the mixed moment
Battle array estimated value separates the time-domain signal of the time-domain signal of basket vibration and core vibration from vibration mixed signal, by
This has achieved the purpose that separate basket vibration signal and core vibration signal from tank surface vibration signal, to real
Effective monitoring to the respective working condition of winding and iron core is showed.
In addition, it is necessary to which explanation, the derivation of expression formula corresponding with above-mentioned hybrid matrix estimated value are as follows：
The vibration of winding and iron core by internal firmware structure, oil be transferred to oil tank wall after, at oil tank wall measuring point
The oscillating component υ contributedWAnd υCIt is finally mixed in oil tank wall and forms oil tank of transformer wall vibrations, is i.e. oil tank wall positionPlace
Vibration can be written as：
υT(xT| t)=υW(xT|t)+υC(xT|t), (1)
According to the mechanism of winding and iron coring vibration, formula (1) can be rewritten as：
Wherein,WithIndicate winding and core vibration excitation point position to oil tank wall response point respectivelyThe overall equivalent transfer function at place.Symbol * then indicates convolution algorithm.F1(i (t)) is winding in excitation point xWThe unit electricity at place
Dynamic distributed forceIn only with the relevant parts electric current i (t), and F2(u (t)) then indicates iron core in excitation point xCThe list at place
Position magnetostrictive forceIn only with the relevant parts voltage u (t).
It is chosen at a certain point position xTDifferent short time period Δ t were in one day1With Δ t2Two groups of vibrations of interior acquisition are believed
Number sampleWithObservation signal as mixed model.According to above-mentioned formula (2), it is known that：
Wherein, the current relationship in different time sections can be considered WithFor different time
Section Δ t1With Δ t2Interior current effective value.When above formula meets above-mentioned mixed signal screening conditions, the winding in different time sections
Relationship between electronic distributed force meets：
Wherein
In addition, when voltage magnitude keeps approximate constant, according to formula (2) it is found that the variable quantity of iron coring vibration component
It is approximately zero, therefore, the distributed force of iron core meets following formula：
And meet
Wherein b22≈ 1,WithRespectively different time sections Δ t1With Δ t2Interior voltage effective value.
In conclusion the discrete form of the mixed model of the transformer vibration based on the vibration of single measuring point may be defined as：
Furthermore, it is contemplated that the Vibration signal of different moments is on the influence that may be present of signal source phase, therefore by the time
Lagged variable μljAbove-mentioned formula (6) is introduced, is obtained：
Wherein l=1,2；J=1,2.So far, it is blind to be successfully converted into traditional signal for the vibration mixed model in transformer
Disjunctive model.Above-mentioned (7) formula is subjected to Fourier transformation, can be obtained：
It is write above formula (8) as matrix form, is obtained：
V (ω)=B (ω) S (ω), (9)
Wherein, S (ω)=[S1(ω),S2(ω)]T, V (ω)=[V1(ω),V2(ω)]T, on this basis and then obtain
Expression formula corresponding with above-mentioned hybrid matrix estimated value, i.e.,：
According to above-mentioned formula (9) it is found that S15 obtains hybrid matrix estimated value B (ω) through the above steps after, in order to acquire
S (ω) need to convert above-mentioned formula (9), i.e.,：
S (ω)=B-1(ω)V(ω), (10)
Wherein, B-1(ω) is the inverse matrix obtained after inverting to above-mentioned hybrid matrix estimated value B (ω)；Due to S
(ω) is the Fourier transformation of basket vibration source signal and core vibration source signal as a result, i.e. S (ω) contains basket vibration source
Frequency-region signal corresponding to signal and core vibration source signal, time-domain signal, need to carry out anti-Fourier's change to S (ω) in order to obtain
It changes, to obtain the time-domain signal of basket vibration and the time-domain signal of core vibration.
The embodiment of the invention discloses a kind of specifically power transformer Separation by vibration method based on time-frequency than distribution, phases
For a upper embodiment, the present embodiment has made further instruction and optimization to technical solution.Specifically：
The detailed process of upper embodiment step S13 includes：By when frequency ratio calculation formula, calculate first group of mixed signal
When frequency ratio between second group of mixed signal；Wherein, frequency ratio calculation formula is when：
Wherein, V1(n, ω) is the Short Time Fourier Transform of first group of mixed signal；V2(n, ω) is second group of mixed signal
Short time discrete Fourier transform；S1(n, ω) is basket vibration source signal s1(n) Short Time Fourier Transform；S2(n, ω) is iron core
Vibrate source signal s2(n) Short Time Fourier Transform.
In upper embodiment step S14, the process of the amplitude best estimate of frequency ratio when calculating, including：
Step S141：Using Dan Yuan time-frequencies area prescreening condition, pre- menu source time-frequency is filtered out from vibration mixed signal
Area；Wherein, Dan Yuan time-frequencies area prescreening condition is：| | β (n, ω) |-| β (n+1, ω) | | < θ1；Wherein, | β (n, ω) | it indicates
When frequency ratio amplitude；θ1For pre-set first threshold；
Step S142：Clustering is carried out to the when frequency ratio amplitude in pre- menu source time-frequency area, is built for characterizing amplitude
The distribution function of distributed degrees；Wherein, distribution function is：
Step S143：The local maximum for calculating distribution function, obtains the first maximum and the second maximum；
Step S144：The tangent value for calculating separately the first maximum and the second maximum is correspondingly made available the first tangent value
With the second tangent value, the amplitude best estimate b of frequency ratio when the first tangent value is determined as21, when the second tangent value is determined as
The amplitude best estimate b of frequency ratio22；
Wherein, feature vector
Position vector
It should be noted that in the present embodiment, the definition in Dan Yuan time-frequencies area is：When in a certain time-frequency area (n, ω) there is only
Single source signal (i.e. there is only basket vibration source signals, or there is only core vibration source signals) or another source signal
Energy very little (is less than preset energy threshold), then the time-frequency area (n, ω) is defined as Dan Yuan time-frequencies area.According to above-mentioned list
The definition in source time-frequency area and it is above-mentioned when frequency ratio calculation formula it is found that in Dan Yuan time-frequencies area, when frequency ratio amplitude | β (n, ω) | be
The expression formula of one constant, the constant isWherein, j is 1 or 2.
In addition, in previous step S14, the process of the phase best estimate of frequency ratio when calculating, including：
Step S145：Using the first Zu Danyuan time-frequencies area screening conditions, filtered out from pre- menu source time-frequency area and winding
Vibrate source signal s1(n) corresponding first Zu Danyuan time-frequencies area SSW1；
Step S146：Using the second Zu Danyuan time-frequencies area screening conditions, filtered out from pre- menu source time-frequency area and iron core
Vibrate source signal s2(n) corresponding second Zu Danyuan time-frequencies area SSW2；
Step S147：Calculate separately the first Zu Danyuan time-frequencies area SSW1With the second Zu Danyuan time-frequencies area SSW2Phase mean value,
It is correspondingly made available first phase mean value and second phase mean value, the phase best estimate of frequency ratio when first phase mean value is determined as
Value μ21, the phase best estimate μ of frequency ratio when second phase mean value is determined as22；
Wherein, the first Zu Danyuan time-frequencies area screening conditions are：||β(n,ω)|-b21| < θ2, the second Zu Danyuan time-frequencies area sieve
The condition is selected to be：||β(n,ω)|-b22| < θ2, θ2For pre-set second threshold；
The calculation formula of first phase mean value is：Wherein,For first group of list source when
Frequency area SSW1When frequency ratio phase value, N1For the first Zu Danyuan time-frequencies area SSW1The number in Zhong Danyuan time-frequencies area；
The calculation formula of second phase mean value is：Wherein,For second group of list source when
Frequency area SSW2When frequency ratio phase value, N2For the second Zu Danyuan time-frequencies area SSW2The number in Zhong Danyuan time-frequencies area.
The embodiment of the invention also discloses a kind of power transformer Separation by vibration systems based on time-frequency than distribution, referring to figure
Shown in 2, which includes：
Signal pickup assembly 21, the vibration signal for the tank surface to power transformer are acquired, are vibrated
Mixed signal；
Signal screening module 22 is filtered out for utilizing preset mixed signal screening conditions from vibration mixed signal
First group of mixed signal and second group of mixed signal；
When frequency ratio computing module 23, for calculating the when frequency ratio between first group of mixed signal and second group of mixed signal；
Best estimate computing module 24, when for calculating separately the amplitude best estimate of frequency ratio and when frequency ratio phase
Best estimate；
Matrix construction module 25, for utilizing amplitude best estimate and phase best estimate, construction hybrid matrix to estimate
Evaluation, and invert to hybrid matrix estimated value, obtain corresponding inverse matrix；
The result accordingly obtained is multiplied by by signal separation module 26 for carrying out Fourier transform to vibration mixed signal
Inverse matrix obtains the Fourier transformation of basket vibration source signal and core vibration source signal as a result, and by basket vibration source signal
Inversefouriertransform is carried out with the Fourier transform results of core vibration source signal, is correspondingly made available the time-domain signal of basket vibration
With the time-domain signal of core vibration；
Wherein, the expression formula of hybrid matrix estimated value is：
Wherein, b21For with basket vibration source signal s1(n) corresponding amplitude best estimate, b22To believe with core vibration source
Number s2(n) corresponding amplitude best estimate, μ21For with basket vibration source signal s1(n) corresponding phase best estimate, μ22
For with core vibration source signal s2(n) corresponding phase best estimate.
In the present embodiment, above-mentioned mixed signal screening conditions are：Voltage effective value and second in first group of mixed signal
The absolute value of the difference of voltage effective value in group mixed signal is less than preset voltage value, and the electric current in first group of mixed signal
The absolute value of the difference of virtual value and the current effective value in second group of mixed signal is more than pre-set current value.
Specifically, the when frequency ratio computing module 23 in the present embodiment, be specifically used for by when frequency ratio calculation formula, calculate the
When frequency ratio between one group of mixed signal and second group of mixed signal；Wherein, frequency ratio calculation formula is when：
Wherein, V1(n, ω) is the Short Time Fourier Transform of first group of mixed signal；V2(n, ω) is second group of mixed signal
Short time discrete Fourier transform；S1(n, ω) is basket vibration source signal s1(n) Short Time Fourier Transform；S2(n, ω) is iron core
Vibrate source signal s2(n) Short Time Fourier Transform.
Shown in Figure 3, in the present embodiment, best estimate computing module 24 can specifically include amplitude best estimate
Computing module 241 and phase best estimate computing module 242.
Wherein, amplitude best estimate computing module 241 specifically includes Dan Yuan time-frequencies area prescreening unit, distribution function structure
Make unit, local maximum computing unit and tangent value computing unit；Wherein,
Dan Yuan time-frequencies area prescreening unit is sieved for utilizing Dan Yuan time-frequencies area prescreening condition from vibration mixed signal
Select pre- menu source time-frequency area；Wherein, Dan Yuan time-frequencies area prescreening condition is：| | β (n, ω) |-| β (n+1, ω) | | < θ1；Its
In, | β (n, ω) | the amplitude of frequency ratio when expression；θ1For pre-set first threshold；
Distribution function structural unit, for carrying out clustering, structure to the when frequency ratio amplitude in pre- menu source time-frequency area
Distribution function for characterizing amplitude distribution degree；Wherein, distribution function is：
Local maximum computing unit, the local maximum for calculating distribution function obtain the first maximum and second
Maximum；
Tangent value computing unit, the tangent value for calculating separately the first maximum and the second maximum, is correspondingly made available
First tangent value and the second tangent value, the amplitude best estimate b of frequency ratio when the first tangent value is determined as21, by the second tangent
The amplitude best estimate b of frequency ratio when value is determined as22；
Wherein, feature vector
Position vector
Secondly, phase best estimate computing module 242 specifically include the first Zu Danyuan time-frequencies area screening unit, second group
Dan Yuan time-frequencies area's screening unit and phase average calculation unit；Wherein,
First Zu Danyuan time-frequencies area screening unit, for utilizing the first Zu Danyuan time-frequencies area screening conditions, from pre- menu source
It is filtered out in time-frequency area and basket vibration source signal s1(n) corresponding first Zu Danyuan time-frequencies area SSW1；
Second Zu Danyuan time-frequencies area screening unit, for utilizing the second Zu Danyuan time-frequencies area screening conditions, from pre- menu source
It is filtered out in time-frequency area and core vibration source signal s2(n) corresponding second Zu Danyuan time-frequencies area SSW2；
Phase average calculation unit, for calculating separately the first Zu Danyuan time-frequencies area SSW1With the second Zu Danyuan time-frequencies area
SSW2Phase mean value, be correspondingly made available first phase mean value and second phase mean value, first phase mean value be determined as time-frequency
The phase best estimate μ of ratio21, the phase best estimate μ of frequency ratio when second phase mean value is determined as22；
Wherein, the first Zu Danyuan time-frequencies area screening conditions are：||β(n,ω)|-b21| < θ2, the second Zu Danyuan time-frequencies area sieve
The condition is selected to be：||β(n,ω)|-b22| < θ2, θ2For pre-set second threshold；
The calculation formula of first phase mean value is：Wherein,For first group of list source when
Frequency area SSW1When frequency ratio phase value, N1For the first Zu Danyuan time-frequencies area SSW1The number in Zhong Danyuan time-frequencies area；
The calculation formula of second phase mean value is：Wherein,For second group of list source when
Frequency area SSW2When frequency ratio phase value, N2For the second Zu Danyuan time-frequencies area SSW2The number in Zhong Danyuan time-frequencies area.
As it can be seen that the embodiment of the present invention filters out first group of mixed signal and second group of mixed signal from vibration mixed signal
Afterwards, the when frequency ratio between first group of mixed signal and second group of mixed signal is just calculated, when by calculating this corresponding to frequency ratio
Amplitude best estimate and phase best estimate, and then corresponding hybrid matrix estimated value is built, then utilize the mixed moment
Battle array estimated value separates the time-domain signal of the time-domain signal of basket vibration and core vibration from vibration mixed signal, by
This has achieved the purpose that separate basket vibration signal and core vibration signal from tank surface vibration signal, to real
Effective monitoring to the respective working condition of winding and iron core is showed.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only that
A little elements, but also include other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Above to a kind of power transformer Separation by vibration method and system based on time-frequency than distribution provided by the present invention
It is described in detail, principle and implementation of the present invention are described for specific case used herein, the above reality
The explanation for applying example is merely used to help understand the method and its core concept of the present invention；Meanwhile for the general technology of this field
Personnel, according to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion this theory
Bright book content should not be construed as limiting the invention.
Claims (10)
1. a kind of power transformer Separation by vibration method based on time-frequency than distribution, which is characterized in that including：
The vibration signal of the tank surface of power transformer is acquired, vibration mixed signal is obtained；
Using preset mixed signal screening conditions, first group of mixed signal and second are filtered out from the vibration mixed signal
Group mixed signal；
Calculate the when frequency ratio between first group of mixed signal and second group of mixed signal；
When calculating separately described the amplitude best estimate of frequency ratio and it is described when frequency ratio phase best estimate；
Using the amplitude best estimate and the phase best estimate, hybrid matrix estimated value is constructed, and to described mixed
It closes Matrix Estimation value to invert, obtains corresponding inverse matrix；
Fourier transformation is carried out to vibration mixed signal, the result accordingly obtained is multiplied by the inverse matrix, obtains basket vibration
The Fourier transformation of source signal and core vibration source signal is as a result, and by the basket vibration source signal and core vibration source signal
Fourier transformation result carry out inversefouriertransform, be correspondingly made available the time-domain signal of basket vibration and the time domain of core vibration
Signal；
Wherein, the expression formula of the hybrid matrix estimated value is：
Wherein, b21For with basket vibration source signal s1(n) corresponding amplitude best estimate, b22For with core vibration source signal s2
(n) corresponding amplitude best estimate, μ21For with the basket vibration source signal s1(n) corresponding phase best estimate, μ22
For with the core vibration source signal s2(n) corresponding phase best estimate.
2. the power transformer Separation by vibration method according to claim 1 based on time-frequency than distribution, which is characterized in that institute
Stating mixed signal screening conditions is：
The difference of voltage effective value in first group of mixed signal and the voltage effective value in second group of mixed signal
Absolute value is less than preset voltage value, and the current effective value in first group of mixed signal and second group of mixed signal
In current effective value absolute value of the difference be more than pre-set current value.
3. the power transformer Separation by vibration method according to claim 2 based on time-frequency than distribution, which is characterized in that institute
State calculate between first group of mixed signal and second group of mixed signal when frequency ratio process, including：
By when frequency ratio calculation formula, calculate the time-frequency between first group of mixed signal and second group of mixed signal
Than；Wherein, frequency ratio calculation formula is when described：
Wherein, V1(n, ω) is the Short Time Fourier Transform of first group of mixed signal；V2(n, ω) is second group of mixing
The Short Time Fourier Transform of signal；S1(n, ω) is the basket vibration source signal s1(n) Short Time Fourier Transform；S2(n,
ω) it is the core vibration source signal s2(n) Short Time Fourier Transform.
4. the power transformer Separation by vibration method according to claim 3 based on time-frequency than distribution, which is characterized in that meter
The process of the amplitude best estimate of frequency ratio when calculating described, including：
Using Dan Yuan time-frequencies area prescreening condition, pre- menu source time-frequency area is filtered out from the vibration mixed signal；Wherein, institute
Shu Danyuan time-frequencies area prescreening condition is：| | β (n, ω) |-| β (n+1, ω) | | < θ1；Wherein, | β (n, ω) | frequency ratio when expression
Amplitude；θ1For pre-set first threshold；
Clustering is carried out to the when frequency ratio amplitude in the pre- menu source time-frequency area, is built for characterizing amplitude distribution degree
Distribution function；Wherein, the distribution function is：
The local maximum for calculating the distribution function obtains the first maximum and the second maximum；
The tangent value for calculating separately first maximum and second maximum, is correspondingly made available the first tangent value and second
Tangent value, the amplitude best estimate b of frequency ratio when first tangent value is determined as21, second tangent value is determined as
When frequency ratio amplitude best estimate b22；
Wherein, SSW indicates Dan Yuan time-frequencies area, feature vector
Position vector
5. the power transformer Separation by vibration method according to claim 4 based on time-frequency than distribution, which is characterized in that meter
The process of the phase best estimate of frequency ratio when calculating described, including：
Using the first Zu Danyuan time-frequencies area screening conditions, filtered out and the basket vibration source from the pre- menu source time-frequency area
Signal s1(n) corresponding first Zu Danyuan time-frequencies area SSW1；
Using the second Zu Danyuan time-frequencies area screening conditions, filtered out and the core vibration source from the pre- menu source time-frequency area
Signal s2(n) corresponding second Zu Danyuan time-frequencies area SSW2；
Calculate separately the first Zu Danyuan time-frequencies area SSW1With the second Zu Danyuan time-frequencies area SSW2Phase mean value, accordingly
Ground obtains first phase mean value and second phase mean value, the phase best estimate of frequency ratio when the first phase mean value is determined as
Value μ21, the phase best estimate μ of frequency ratio when the second phase mean value is determined as22；
Wherein, the first Zu Danyuan time-frequencies area screening conditions are：||β(n,ω)|-b21| < θ2, second group of list source time-frequency
Area's screening conditions are：||β(n,ω)|-b22| < θ2, θ2For pre-set second threshold；
The calculation formula of the first phase mean value is：Wherein,For first group of list source
Time-frequency area SSW1When frequency ratio phase value, N1For the first Zu Danyuan time-frequencies area SSW1The number in Zhong Danyuan time-frequencies area；
The calculation formula of the second phase mean value is：Wherein,For second group of list
Source time-frequency area SSW2When frequency ratio phase value, N2For the second Zu Danyuan time-frequencies area SSW2The number in Zhong Danyuan time-frequencies area.
6. a kind of power transformer Separation by vibration system based on time-frequency than distribution, which is characterized in that including：
Signal pickup assembly, the vibration signal for the tank surface to power transformer are acquired, and obtain vibration mixing letter
Number；
Signal screening module filters out for utilizing preset mixed signal screening conditions from the vibration mixed signal
One group of mixed signal and second group of mixed signal；
When frequency ratio computing module, for calculating the time-frequency between first group of mixed signal and second group of mixed signal
Than；
Best estimate computing module, when for calculating separately described the amplitude best estimate of frequency ratio and it is described when frequency ratio phase
Position best estimate；
Matrix construction module constructs hybrid matrix for utilizing the amplitude best estimate and the phase best estimate
Estimated value, and invert to the hybrid matrix estimated value, obtain corresponding inverse matrix；
The result accordingly obtained is multiplied by described inverse by signal separation module for carrying out Fourier transformation to vibration mixed signal
Matrix obtains the Fourier transformation of basket vibration source signal and core vibration source signal as a result, and believing the basket vibration source
Number and core vibration source signal Fourier transformation result carry out inversefouriertransform, be correspondingly made available basket vibration time domain letter
Number and core vibration time-domain signal；
Wherein, the expression formula of the hybrid matrix estimated value is：
Wherein, b21For with basket vibration source signal s1(n) corresponding amplitude best estimate, b22For with core vibration source signal s2
(n) corresponding amplitude best estimate, μ21For with the basket vibration source signal s1(n) corresponding phase best estimate, μ22
For with the core vibration source signal s2(n) corresponding phase best estimate.
7. the power transformer Separation by vibration system according to claim 6 based on time-frequency than distribution, which is characterized in that institute
Stating mixed signal screening conditions is：
The difference of voltage effective value in first group of mixed signal and the voltage effective value in second group of mixed signal
Absolute value is less than preset voltage value, and the current effective value in first group of mixed signal and second group of mixed signal
In current effective value absolute value of the difference be more than pre-set current value.
8. the power transformer Separation by vibration system according to claim 7 based on time-frequency than distribution, which is characterized in that
Frequency ratio computing module when described, frequency ratio calculation formula when being specifically used for passing through calculate first group of mixed signal and institute
State the when frequency ratio between second group of mixed signal；Wherein, frequency ratio calculation formula is when described：
Wherein, V1(n, ω) is the Short Time Fourier Transform of first group of mixed signal；V2(n, ω) is second group of mixing
The Short Time Fourier Transform of signal；S1(n, ω) is the basket vibration source signal s1(n) Short Time Fourier Transform；S2(n,
ω) it is the core vibration source signal s2(n) Short Time Fourier Transform.
9. the power transformer Separation by vibration system according to claim 8 based on time-frequency than distribution, which is characterized in that institute
It includes amplitude best estimate computing module and phase best estimate computing module to state best estimate computing module；Wherein,
The amplitude best estimate computing module includes：
Dan Yuan time-frequencies area prescreening unit is sieved for utilizing Dan Yuan time-frequencies area prescreening condition from the vibration mixed signal
Select pre- menu source time-frequency area；Wherein, Dan Yuan time-frequencies area prescreening condition is：| | β (n, ω) |-| β (n+1, ω) | | <
θ1；Wherein, | β (n, ω) | the amplitude of frequency ratio when expression；θ1For pre-set first threshold；
Distribution function structural unit, for carrying out clustering, structure to the when frequency ratio amplitude in the pre- menu source time-frequency area
Distribution function for characterizing amplitude distribution degree；Wherein, the distribution function is：
Local maximum computing unit, the local maximum for calculating the distribution function obtain the first maximum and second
Maximum；
Tangent value computing unit, the tangent value for calculating separately first maximum and second maximum, correspondingly
The first tangent value and the second tangent value are obtained, the amplitude best estimate b of frequency ratio when first tangent value is determined as21, will
The amplitude best estimate b of frequency ratio when second tangent value is determined as22；
Wherein, feature vector
Position vector
10. the power transformer Separation by vibration system according to claim 9 based on time-frequency than distribution, which is characterized in that
The phase best estimate computing module includes：
First Zu Danyuan time-frequencies area screening unit, for utilizing the first Zu Danyuan time-frequencies area screening conditions, from the pre- menu source
It is filtered out in time-frequency area and the basket vibration source signal s1(n) corresponding first Zu Danyuan time-frequencies area SSW1；
Second Zu Danyuan time-frequencies area screening unit, for utilizing the second Zu Danyuan time-frequencies area screening conditions, from the pre- menu source
It is filtered out in time-frequency area and the core vibration source signal s2(n) corresponding second Zu Danyuan time-frequencies area SSW2；
Phase average calculation unit, for calculating separately the first Zu Danyuan time-frequencies area SSW1With second group of list source time-frequency
Area SSW2Phase mean value, be correspondingly made available first phase mean value and second phase mean value, by the first phase mean value determine
For when frequency ratio phase best estimate μ21, the phase best estimate μ of frequency ratio when the second phase mean value is determined as22；
Wherein, the first Zu Danyuan time-frequencies area screening conditions are：||β(n,ω)|-b21| < θ2, second group of list source time-frequency
Area's screening conditions are：||β(n,ω)|-b22| < θ2, θ2For pre-set second threshold；
The calculation formula of the first phase mean value is：Wherein,For first group of list source
Time-frequency area SSW1When frequency ratio phase value, N1For the first Zu Danyuan time-frequencies area SSW1The number in Zhong Danyuan time-frequencies area；
The calculation formula of the second phase mean value is：Wherein,For second group of list
Source time-frequency area SSW2When frequency ratio phase value, N2For the second Zu Danyuan time-frequencies area SSW2The number in Zhong Danyuan time-frequencies area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511030951.4A CN105527016B (en) | 2015-12-31 | 2015-12-31 | A kind of power transformer Separation by vibration method and system based on time-frequency than distribution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511030951.4A CN105527016B (en) | 2015-12-31 | 2015-12-31 | A kind of power transformer Separation by vibration method and system based on time-frequency than distribution |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105527016A CN105527016A (en) | 2016-04-27 |
| CN105527016B true CN105527016B (en) | 2018-11-13 |
Family
ID=55769385
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511030951.4A Active CN105527016B (en) | 2015-12-31 | 2015-12-31 | A kind of power transformer Separation by vibration method and system based on time-frequency than distribution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105527016B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107941324A (en) * | 2017-10-08 | 2018-04-20 | 天津大学 | A kind of method of estimation of consumer level inertia sensing unit environments noise |
| CN108520115B (en) * | 2018-03-22 | 2022-02-18 | 国网湖南省电力有限公司 | Method and system for separating vibration signals of transformer winding and iron core |
| CN110658398B (en) * | 2019-09-16 | 2021-07-13 | 西安交通大学 | Transformer vibration fundamental frequency signal separation method based on power factor angle correction |
| CN114113939B (en) * | 2021-11-24 | 2022-11-18 | 国网宁夏电力有限公司宁东供电公司 | Transient vibration signal-based transformer fault detection method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101825489A (en) * | 2010-01-29 | 2010-09-08 | 浙江大学 | Method for separating OLTC (On-Load Tap Changer) vibration signals of power transformer |
| CN102914362A (en) * | 2012-10-12 | 2013-02-06 | 宁夏电力公司电力科学研究院 | Separating method for vibration signals of transformer winding and iron core |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101260684B1 (en) * | 2011-05-30 | 2013-05-10 | 서강대학교산학협력단 | Blind source separation method and apparatus according to independent vector analysis using feed forward network |
-
2015
- 2015-12-31 CN CN201511030951.4A patent/CN105527016B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101825489A (en) * | 2010-01-29 | 2010-09-08 | 浙江大学 | Method for separating OLTC (On-Load Tap Changer) vibration signals of power transformer |
| CN102914362A (en) * | 2012-10-12 | 2013-02-06 | 宁夏电力公司电力科学研究院 | Separating method for vibration signals of transformer winding and iron core |
Non-Patent Citations (2)
| Title |
|---|
| Time-frequency Sparsity Map on Automatic Partial Discharge Sources Separation for Power Transformer Condition Assessment;Jeffery C.Chan等;《IEEE Transactions on Dielectrics and Electrical Insulation》;20150831;第22卷(第4期);第2271-2283页 * |
| 基于独立成分分析和端点检测的变压器有载分接开关振动信号自适应分离;郑婧等;《电网技术》;20101130;第34卷(第11期);第208-213页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105527016A (en) | 2016-04-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105527016B (en) | A kind of power transformer Separation by vibration method and system based on time-frequency than distribution | |
| Wang et al. | Application of computed order tracking, Vold–Kalman filtering and EMD in rotating machine vibration | |
| Luo et al. | A kurtosis-guided adaptive demodulation technique for bearing fault detection based on tunable-Q wavelet transform | |
| Coenen et al. | A multi-scale approach to bridge microscale damage and macroscale failure: a nested computational homogenization-localization framework | |
| Makki Alamdari et al. | Spectral-based damage identification in structures under ambient vibration | |
| CN102753982B (en) | Method and system of subsynchronous resonance detection | |
| Bao et al. | Multi-stage identification scheme for detecting damage in structures under ambient excitations | |
| Le et al. | Distinction between harmonic and structural components in ambient excitation tests using the time–frequency domain decomposition technique | |
| CN104698325B (en) | A kind of method of discrimination of the low-frequency oscillation of power system negative damping mechanism and forced oscillation | |
| CN107423261A (en) | The separation method of positive and negative order components based on OVPR under the conditions of non-ideal micro-capacitance sensor | |
| CN105891625B (en) | Subsynchronous oscillation of electrical power system disturbing source discriminating conduct based on energy stream | |
| Kumar et al. | Viscoelasticity of coagulated alumina suspensions | |
| CN110174255A (en) | A kind of transformer vibration signal separation method based on radial base neural net | |
| Meng et al. | Adaptive sparse denoising and periodicity weighted spectrum separation for compound bearing fault diagnosis | |
| Mountrakis et al. | Validation of an efficient two-dimensional model for dense suspensions of red blood cells | |
| Wu et al. | Weak signal enhancement by fractional-order system resonance and its application in bearing fault diagnosis | |
| Kelley et al. | Adaptive synchronized switch damping on an inductor: a self-tuning switching law | |
| Bucher et al. | A material model for finite elasto-plastic deformations considering a substructure | |
| CN104678170B (en) | A kind of Electric Power Harmonic Analysis method harmonic analyzer based on harmonic analysis instrument | |
| Yanez-Borjas et al. | Nonlinear mode decomposition-based methodology for modal parameters identification of civil structures using ambient vibrations | |
| CN211293127U (en) | Partial discharge measurement system under harmonic voltage effect | |
| CN106596025A (en) | Highway tunnel hanging fan base stability detection method and system based on impulse response | |
| Shao et al. | Simultaneous de-noising and enhancement method for long-span bridge health monitoring data based on empirical mode decomposition and fractal conservation law | |
| CN105911153A (en) | Signal separating and denoising method and apparatus based on moving window function | |
| Cheng et al. | A combined crosstalk cancellation method based on wavelet packet denoising and Welch’s method for operational transfer path analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |