CN107741271A - A kind of Winding in Power Transformer state evaluating method based on system delay Order- reduction - Google Patents
A kind of Winding in Power Transformer state evaluating method based on system delay Order- reduction Download PDFInfo
- Publication number
- CN107741271A CN107741271A CN201710867458.0A CN201710867458A CN107741271A CN 107741271 A CN107741271 A CN 107741271A CN 201710867458 A CN201710867458 A CN 201710867458A CN 107741271 A CN107741271 A CN 107741271A
- Authority
- CN
- China
- Prior art keywords
- mrow
- winding
- msup
- power transformer
- msub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Housings And Mounting Of Transformers (AREA)
Abstract
The invention discloses a kind of Winding in Power Transformer state evaluating method based on system delay Order- reduction, this method is based on basket vibration mechanism of production, preferably reflect the mechanical structure state of winding, winding failure is directly associated with characteristic quantity, foundation is provided with scientific for the validity of fault diagnosis.In addition, the present invention is implemented with transformer without any electrical connection, and without be powered off to transformer, to the influence on system operation very little of whole power system.
Description
Technical field
The invention belongs to power transformer safety failure detection technique field, and in particular to one kind is based on system delay exponent number
The Winding in Power Transformer state evaluating method of estimation.
Background technology
Large-scale power transformer as a highly important ring in power system, its safe operation to ensure power grid security,
Reliable power supply holds the balance.In Power Transformer Faults, the defective proportion of winding is up to 46.4%, for transformer fault most
Critical piece, and catastrophe failure accounts for the 70% of winding total failare caused by winding generation mechanically deform under electrodynamic action.
The even slight deformation of winding, the problems such as also triggering winding machinery penalty, dielectric strength and anti-short circuit capability to reduce,
Bring great potential safety hazard.Therefore, the status monitoring to its major failure part winding is realized in transformer belt pyroelectric monitor
It is very necessary and important with assessment.
Vibration analysis method obtains inside transformer unit status by analyzing the oil tank wall vibration signal collected
Information.When inside transformer mechanical structure changes, the system of vibrational system certainly will be caused to respond and changed.Therefore,
Input electromagnetic force can be built with exporting the relational model of basket vibration response, and the system performance reflection by extracting the model becomes
The mechanical structure state of depressor winding.
The content of the invention
In view of above-mentioned, the invention provides a kind of Winding in Power Transformer state estimation based on system delay Order- reduction
Method, can on-line checking go out Winding in Power Transformer mechanical structure state.
A kind of Winding in Power Transformer state evaluating method based on system delay Order- reduction, comprises the following steps：
(1) correspond to winding position in power transformer tank surface and dispersedly arrange multiple vibrating sensors, record electricity
The vibration signal of transformer each vibrating sensor under low-voltage load operating conditions, and synchronous acquisition power transformer is once
Side electric current；
(2) vibration signal and current signal are pre-processed, magnetic hysteresis is gone including to current signal
Processing；
(3) for any vibrating sensor, passed by building the relational model of electric current and basket vibration, and according to the vibration
The pretreated vibration signal of sensor and current signal estimate characteristic quantities of the delay exponent number n of model as system performance, n
=Na+Nb, NaAnd NbThe respectively actual linear delay exponent number of system output and input, NaAnd NbIt is natural number and Na=NbOr
Na=Nb+1；And then according to n, NaAnd NbJudge the machinery knot based on Winding in Power Transformer under the vibrating sensor data cases
Structure state；
(4) all vibrating sensors are traveled through according to step (3), when judging winding based on a certain proportion of vibrating sensor
Mechanical structure state is normal, then finally judges that Winding in Power Transformer is normal, otherwise judges that Winding in Power Transformer is abnormal.
Further, the detailed process pre-processed in the step (2) to vibration signal and current signal is：It is first
First, vibration signal and current signal are normalized；Then, the current signal after normalization is entered according to below equation
Row nonlinear transformation：
Wherein：I (t) is the current signal after normalization, and t is the moment, ip(t) it is the current signal after nonlinear transformation.
Further, appraising model delay exponent number n detailed process is as follows in the step (3)：
3.1 according to below equation computing relay exponent number be n in the case of in relational model all sampled points combine
Lipschitz coefficients：
Wherein：lij (n)Represent what delay exponent number combined for the i-th sampled point in relational model in the case of n with jth sampled point
Lipschitz coefficients, i and j are sampled point sequence number；When n is even number, γ1(i)~γn(i) it is corresponding to be equal to y (i-1), x
(i-1),y(i-2),x(i-2),...,y(i-Na),x(i-Nb), γ1(j)~γn(j) it is corresponding to be equal to y (j-1), x (j-1), y
(j-2), x(j-2),...,y(j-Na),x(j-Nb)；When n is odd number, γ1(i)~γn(i) it is corresponding to be equal to y (i-1), x
(i-1),y(i-2), x(i-2),...,y(i-Nb),x(i-Nb),y(i-Na), γ1(j)~γn(j) it is corresponding to be equal to y (j-1), x
(j-1),y(j-2), x(j-2),...,y(j-Nb),x(j-Nb),y(j-Na)；Y (i) and y (j) is respectively that vibrating sensor is pre-
The signal value of corresponding i-th sampled point and jth sampled point in vibration signal after processing, y (i-1) and y (j-1) are respectively to vibrate to pass
The signal value of corresponding i-th -1 sampled point and the sampled point of jth -1, y (i-2) and y (j-2) divide in the pretreated vibration signal of sensor
Not Wei in the pretreated vibration signal of vibrating sensor corresponding i-th -2 sampled point and the sampled point of jth -2 signal value, y (i-
Na) and y (j-Na) it is respectively corresponding i-th-N in the pretreated vibration signal of vibrating sensoraSampled point and jth-NaSampling
The signal value of point, y (i-Nb) and y (j-Nb) it is respectively corresponding i-th-N in the pretreated vibration signal of vibrating sensorbSampling
Point and jth-NbThe signal value of sampled point, x (i-1) and x (j-1) are respectively that the i-th -1 sampling is corresponded in the current signal after handling
The signal value of point and the sampled point of jth -1, x (i-2) and x (j-2) are respectively to correspond to the i-th -2 sampled point in the current signal after handling
With the signal value of the sampled point of jth -2, x (i-Nb) and x (j-Nb) it is respectively corresponding i-th-N in current signal after handlingbSampled point
With jth-NbThe signal value of sampled point；
3.2 sort all Lipschitz coefficients being calculated in step 3.1 and m before intercepting from big to small
Lipschitz coefficients, and then according to the Lipschitz averages that below equation computing relay exponent number is relational model in the case of n
Number：
Wherein：l(n)Represent the Lipschitz mean coefficients that delay exponent number is relational model in the case of n, l(n)(z) it is from big
Z-th of Lipschitz coefficient after to small sequence, m generally take 0.01Nset, NsetFor total number of sample points；
3.3, according to step 3.1 and 3.2, make delay exponent number n cumulative traversals since 1, the output pair when meeting following condition
Characteristic quantities of the model delay exponent number n answered as system performance；
Wherein：l(n+1)The Lipschitz mean coefficients that delay exponent number is relational model in the case of n+1 are represented, ε is setting
Convergence threshold.
Further, the detailed process that Winding in Power Transformer mechanical structure state is judged in the step (3) is：It is first
First, ensure Winding in Power Transformer calculated under normal circumstances according to step (1) to (3) process the delay exponent number n of model with
And NaAnd Nb；Then, make in step (3) based on n, the N for actually calculating to obtain in the case of current vibration sensing dataaAnd NbWith
N, the N of Winding in Power Transformer under normal circumstancesaAnd NbIt is compared, if both data are coincide, judges to pass based on current vibration
The mechanical structure state of Winding in Power Transformer is normal under sensor data cases；If both data misfit, judgement is based on
The mechanical structure state of Winding in Power Transformer is abnormal in the case of current vibration sensing data.
The advantageous effects of the present invention are as follows：
1. present invention specific implementation and transformer need not power off without any electrical connection to transformer, to whole
The influence on system operation very little of power system.
2. the inventive method is based on basket vibration mechanism of production, the mechanical structure state of winding is preferably reflected, will be around
Group failure is directly associated with characteristic quantity, and foundation is provided with scientific for the validity of fault diagnosis.
Brief description of the drawings
Fig. 1 is the step schematic flow sheet of the inventive method.
Fig. 2 is power transformer tank surface point layout figure.
Fig. 3 is input current i (t) and oscillating componentGraph of relation.
Fig. 4 is the input current i after degaussing hysteresis is changedpAnd oscillating component (t)Graph of relation.
Fig. 5 is input current ip(t) and with oscillating componentFor the delay Order- reduction result figure of output.
Embodiment
In order to more specifically describe the present invention, below in conjunction with the accompanying drawings and embodiment is to technical scheme
It is described in detail.
The specific experiment object of present embodiment is a 110kV three-phase oil-immersed power transformer, in order to verify this hair
The validity of bright method, specially compared for same winding under normal circumstances with the vibration performance in the case of failure.Failure around
Group is that normal winding is realized by short-circuit impact artificial damage, and transformer short-circuit experiment is a kind of test side specifically for winding
Method, by low pressure terminal shortcircuit, apply voltage in high-pressure side, Transformer Winding electric current can be allowed to reach rated value.
, must be higher from sensitivity for the ease of the distortionless vibration signal for obtaining different amplitudes when being diagnosed
Vibrating sensor；In order to ensure vibratory response of the sensor within sampling filter frequency band, vibrating sensor is fixed on oil
Should be by the way of magnetic support absorption or glue bonding in case side wall.Vibrating sampling apparatus includes preposition amplification, anti-aliasing filter, AD
Main modular, wherein AD sampling resolutions at least 12, the frequency overlapped-resistable filter cut-off frequencies such as sampling are 2000Hz；Vibrated
During signal sampling, sample frequency is at least 4000Hz.In the present embodiment, the sample frequency for gathering vibration signal is arranged to
10000Hz, A/D module sampling resolution are 16, and the overall process of experiment is recorded using continuous sampling pattern.
As shown in figure 1, Winding in Power Transformer state evaluating method of the present invention based on systematic education estimation, including it is as follows
Step：
(1) arrangement vibration measuring point.
5 measuring points are arranged in power transformer tank surface, as shown in Fig. 2 because A phase windings can artificially be damaged in experiment
It is bad, so 5 point layouts are corresponding to A phase windings on oil tank wall.
(2) normal condition and vibration signal and current signal under abnormal conditions are gathered.
Transformer new to one first carries out short circuit experiment, steps up the voltage on high-pressure side to increase on winding
Electric current, increase electric current according to 10% ratio every time, progressively reach rated value；After electric current increase by 10%, keep stablizing for 30 seconds it is constant,
Using continuous sampling pattern, the vibration of all measuring point institutes having time is recorded.
Then using short-circuit impact electric current to A phase windings carry out artificial destruction, confirm A phase windings centre position there occurs
Largely deform, then same short circuit experiment is carried out to transformer, equally record all measuring points in whole process
Vibration.
(3) signal is pre-processed.
When with input current i (t) for independent variable, with basket vibration in measuring pointThe oscillating component contributedFor because
(with measuring point when variable draws relation curve between the twoExemplified by, as shown in Figure 3), it is seen that output input relation curve presents obvious
Hysteresis characteristic, this is due to the introduced strong nonlinearity characteristic in magnetic field residing for winding, input signal is carried out it is following non-thread
Property conversion after：
Wherein：I'(t) the differential form for being input signal i (t), ip(t) it is the input signal after conversion.This is non-thread
Property change by original input signal from the low latitude space reflection where it into higher dimensional space, eliminate system Central Plains nonlinear block
In the hysteresis characteristic that carries, Fig. 4 then shown in the input current i after above-mentioned conversionpAnd oscillating component (t)
Relation curve.
(4) relational model delay exponent number is extracted, judges winding machinery configuration state.
From the vibration mechanism of transformer, the electronic distributed force of unit of winding is mainly the electric current by flowing through winding coil
Produce, and relevant with the stray field near winding, the size of stray field is directly related with electric current, but with this body structure of winding and
Winding coil is highly relevant；So as to by the electronic distributed force of unitIt is considered as non-linear letter only related to electric current i (t)
Number F1(i (t)) and nonlinear function only related to coil positionCombination.Therefore, basket vibration and input electricity
The relation formula of stream can be written as：
Wherein：Represent basket vibration exciting force to oil tank wall response pointThe overall EU Equivalent Unit pulse at place rings
Should.
Observe above formula and understand that electric current-basket vibration relational model is by a static non linear module F1() and dynamic linear
ModuleThe nonlinear system of composition, it is consistent with classical Nonlinear Hammerstein model structure, i.e., by an input
Nonlinear Static module forms with a linear dynamic block coupled in series.
From above-mentioned model, the mechanical structure feature of winding will directly affect the nonlinear function related to positionCharacteristic and transmission characteristic, so as to the linear block characteristic that will be directly affected in the model, therefore it is linear to can extract this
Modular system characteristic is as winding machinery structural state amount.
For linear system, it, which postpones exponent number change, will directly affect the response of its system, if therefore can be according to input
Delay exponent number of the output signal feature to the system carries out direct estimation, then using the delay exponent number change to transformer around
Group mechanical structure feature is assessed.
For a nonlinear system, output y (t) can be described as inputting with delay, export relevant function, i.e.,：
Y (t)=g (y (t-1) ..., y (t-Na),x(t-1),…,x(t-Nb))
=g (γ1,γ2,…,γn)
Wherein：X (t) and y (t) is respectively the input and output of nonlinear system, and NaAnd NbRespectively system output and it is defeated
The actual linear delay exponent number entered, g () is a nonlinear function, and assumes that the function has continuity Characteristics, is met
The Lipschitz conditions of continuity.Used here as γi, the mark of i=1 ..., n as the nonlinear function independent variable, and make n=
Na+NbFor independent variable number.
Define Lipschitz coefficients lij (n)To characterize nonlinear function g (γ1,γ2,…,γn) continuity：
When lacking necessary variable γ in function argumenthWhen, then lack the Lipschitz coefficients l during variableij (h-1)Value
It will be far longer than in the presence of the necessary variable γhWhen coefficient value lij (h), i.e., now lij (h-1)＞ ＞ lij (h)；It is on the other side, if
Variable γh+1For a redundancy or inessential independent variable, then its corresponding Lipschitz coefficients lij (h+1)With being during the missing variable
Number lij (h)It is worth approximately equal, i.e. lij (h+1)≈lij (h)。
In order to avoid the influence of the noise that introduces to the above results during measurement, Lipschitz mean coefficients l is used(n)Substitute
Coefficient lij (n):
Wherein：l(n)(z) it is lij (n)By z-th of coefficient value after rule rearrangement of successively decreasing, m is usually using 0.01Nset,
NsetIt is the input-output for Order- reduction to number.
The main purpose of the algorithm is to find minimum independent variable number n*(n*=Na *+Nb *), and make corresponding to it
Lipschitz mean coefficients meet following condition：
Wherein：ε is one by empirical value, and ε=0.1 is used in present embodiment.
Therefore, independent variable number n*Corresponding Na *And Nb *Can be respectively seen as the nonlinear system output y (t) and
Input x (t) minimum delay Order- reduction value.
In an experiment choose 0.5s time spans input and output data carry out linear delay Order- reduction, with normally around
Measuring point under group stateEstimated result is as shown in figure 5, with the increase stepwise of delay exponent number, Lipschitz coefficients l(n)By
Rank declines.As exponent number n >=5, its variation tendency tends towards stability, and more than being met due to corresponding Lipschitz mean coefficients
Convergent requirement.Therefore, it is (3,2) that can obtain linear block Order- reduction value, i.e. Na=3, Nb=2.
Afterwards, all measuring point vibrations under normal winding and abnormal winding state and the delay exponent number of current relationship model are entered
Row estimation, is obtained as shown in the results summarized in table 1：
The Order- reduction result (100% load) of all measuring points under the winding different conditions of table 1.
As known from Table 1, when abnormal (deformation) occurs in winding mechanical structure, electric current-vibration corresponding to most of measuring point
The delay exponent number of relational model can find significant changes, therefore the change that can postpone exponent number by monitoring model is tied to winding machinery
Structure state carries out status monitoring.
The above-mentioned description to embodiment is understood that for ease of those skilled in the art and using the present invention.
Person skilled in the art obviously can easily make various modifications to above-described embodiment, and described herein general
Principle is applied in other embodiment without by performing creative labour.Therefore, the invention is not restricted to above-described embodiment, ability
For field technique personnel according to the announcement of the present invention, the improvement made for the present invention and modification all should be in protection scope of the present invention
Within.
Claims (4)
1. a kind of Winding in Power Transformer state evaluating method based on system delay Order- reduction, comprises the following steps：
(1) correspond to winding position in power transformer tank surface and dispersedly arrange multiple vibrating sensors, record electricity transformation
The vibration signal of device each vibrating sensor under low-voltage load operating conditions, and the primary side electricity of synchronous acquisition power transformer
Stream；
(2) vibration signal and current signal are pre-processed, goes magnetic hysteresisization to handle including to current signal；
(3) for any vibrating sensor, by building the relational model of electric current and basket vibration, and according to the vibrating sensor
Pretreated vibration signal and current signal estimate characteristic quantities of the delay exponent number n of model as system performance, n=Na
+Nb, NaAnd NbThe respectively actual linear delay exponent number of system output and input, NaAnd NbIt is natural number and Na=NbOr Na=
Nb+1；And then according to n, NaAnd NbJudge the mechanical structure shape based on Winding in Power Transformer under the vibrating sensor data cases
State；
(4) all vibrating sensors are traveled through according to step (3), when judging winding machinery based on a certain proportion of vibrating sensor
Configuration state is normal, then finally judges that Winding in Power Transformer is normal, otherwise judges that Winding in Power Transformer is abnormal.
2. Winding in Power Transformer state evaluating method according to claim 1, it is characterised in that：In the step (2)
The detailed process pre-processed to vibration signal and current signal is：First, vibration signal and current signal are returned
One change is handled；Then, nonlinear transformation is carried out to the current signal after normalization according to below equation：
<mrow>
<msub>
<mi>i</mi>
<mi>p</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mi>arccos</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>d</mi>
<mi>i</mi>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>/</mo>
<mi>d</mi>
<mi>t</mi>
<mo>&GreaterEqual;</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mn>2</mn>
<mi>&pi;</mi>
<mo>-</mo>
<mi>arccos</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>d</mi>
<mi>i</mi>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>/</mo>
<mi>d</mi>
<mi>t</mi>
<mo><</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
Wherein：I (t) is the current signal after normalization, and t is the moment, ip(t) it is the current signal after nonlinear transformation.
3. Winding in Power Transformer state evaluating method according to claim 1, it is characterised in that：In the step (3)
Appraising model delay exponent number n detailed process is as follows：
The 3.1 Lipschitz systems combined according to below equation computing relay exponent number for all sampled points in relational model in the case of n
Number：
<mrow>
<msup>
<msub>
<mi>l</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<mfrac>
<mrow>
<mo>|</mo>
<mi>y</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mi>y</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>|</mo>
</mrow>
<msqrt>
<mrow>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>&gamma;</mi>
<mn>1</mn>
</msub>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
<mo>-</mo>
<msub>
<mi>&gamma;</mi>
<mn>1</mn>
</msub>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>&gamma;</mi>
<mn>2</mn>
</msub>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
<mo>-</mo>
<msub>
<mi>&gamma;</mi>
<mn>2</mn>
</msub>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<mn>...</mn>
<mo>+</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>&gamma;</mi>
<mi>n</mi>
</msub>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
<mo>-</mo>
<msub>
<mi>&gamma;</mi>
<mi>n</mi>
</msub>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
</mfrac>
<mo>,</mo>
<mi>i</mi>
<mo>&NotEqual;</mo>
<mi>j</mi>
</mrow>
Wherein：lij (n)Represent what delay exponent number combined for the i-th sampled point in relational model in the case of n with jth sampled point
Lipschitz coefficients, i and j are sampled point sequence number；When n is even number, γ1(i)~γn(i) it is corresponding to be equal to y (i-1), x (i-
1),y(i-2),x(i-2),...,y(i-Na),x(i-Nb), γ1(j)~γn(j) it is corresponding to be equal to y (j-1), x (j-1), y (j-
2),x(j-2),...,y(j-Na),x(j-Nb)；When n is odd number, γ1(i)~γnIt is (i) corresponding to be equal to y (i-1), x (i-1),
y(i-2),x(i-2),...,y(i-Nb),x(i-Nb),y(i-Na), γ1(j)~γn(j) it is corresponding to be equal to y (j-1), x (j-1), y
(j-2),x(j-2),...,y(j-Nb),x(j-Nb),y(j-Na)；Y (i) and y (j) is respectively that vibrating sensor is pretreated
The signal value of corresponding i-th sampled point and jth sampled point in vibration signal, y (i-1) and y (j-1) are respectively that vibrating sensor is located in advance
The signal value of corresponding i-th -1 sampled point and the sampled point of jth -1 in vibration signal after reason, y (i-2) and y (j-2) are respectively to vibrate
The signal value of corresponding i-th -2 sampled point and the sampled point of jth -2, y (i-N in the pretreated vibration signal of sensora) and y (j-
Na) it is respectively corresponding i-th-N in the pretreated vibration signal of vibrating sensoraSampled point and jth-NaThe signal value of sampled point,
y(i-Nb) and y (j-Nb) it is respectively corresponding i-th-N in the pretreated vibration signal of vibrating sensorbSampled point and jth-NbAdopt
The signal value of sampling point, x (i-1) and x (j-1) are respectively that the i-th -1 sampled point and the sampling of jth -1 are corresponded in the current signal after handling
The signal value of point, x (i-2) and x (j-2) are respectively to correspond to the i-th -2 sampled point and the sampled point of jth -2 in the current signal after handling
Signal value, x (i-Nb) and x (j-Nb) it is respectively corresponding i-th-N in current signal after handlingbSampled point and jth-NbSampling
The signal value of point；
3.2 sort all Lipschitz coefficients being calculated in step 3.1 and m Lipschitz before intercepting from big to small
Coefficient, and then according to the Lipschitz mean coefficients that below equation computing relay exponent number is relational model in the case of n：
<mrow>
<msup>
<mi>l</mi>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msup>
<mrow>
<mo>(</mo>
<msqrt>
<mi>n</mi>
</msqrt>
<munderover>
<mo>&Pi;</mo>
<mrow>
<mi>z</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>m</mi>
</munderover>
<msup>
<mi>l</mi>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>(</mo>
<mi>z</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mrow>
<mn>1</mn>
<mo>/</mo>
<mi>m</mi>
</mrow>
</msup>
</mrow>
Wherein：l(n)Represent the Lipschitz mean coefficients that delay exponent number is relational model in the case of n, l(n)(z) for from big to small
Z-th of Lipschitz coefficient after sequence, m generally take 0.01Nset, NsetFor total number of sample points；
3.3, according to step 3.1 and 3.2, make delay exponent number n cumulative traversals since 1, when meeting following condition corresponding to output
Characteristic quantities of the model delay exponent number n as system performance；
<mrow>
<mfrac>
<mrow>
<mo>|</mo>
<mrow>
<msup>
<mi>l</mi>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>+</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msup>
<mo>-</mo>
<msup>
<mi>l</mi>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
</msup>
</mrow>
<mo>|</mo>
</mrow>
<mrow>
<mi>m</mi>
<mi>a</mi>
<mi>x</mi>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>,</mo>
<mo>|</mo>
<msup>
<mi>l</mi>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>|</mo>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo><</mo>
<mi>&epsiv;</mi>
</mrow>
Wherein：l(n+1)The Lipschitz mean coefficients that delay exponent number is relational model in the case of n+1 are represented, ε is the convergence of setting
Threshold value.
4. Winding in Power Transformer state evaluating method according to claim 1, it is characterised in that：In the step (3)
The detailed process for judging Winding in Power Transformer mechanical structure state is：First, Winding in Power Transformer normal condition is being ensured
Lower delay the exponent number n and N that model is calculated according to step (1) to (3) processaAnd Nb；Then, make in step (3) based on current
Obtained n, N is actually calculated under vibrating sensor data casesaAnd NbWith n, the N of Winding in Power Transformer under normal circumstancesaAnd Nb
It is compared, if both data are coincide, judges the machine based on Winding in Power Transformer in the case of current vibration sensing data
Tool configuration state is normal；If both data misfit, judge based on power transformer in the case of current vibration sensing data
The mechanical structure state of device winding is abnormal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710867458.0A CN107741271A (en) | 2017-09-22 | 2017-09-22 | A kind of Winding in Power Transformer state evaluating method based on system delay Order- reduction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710867458.0A CN107741271A (en) | 2017-09-22 | 2017-09-22 | A kind of Winding in Power Transformer state evaluating method based on system delay Order- reduction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107741271A true CN107741271A (en) | 2018-02-27 |
Family
ID=61236159
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710867458.0A Pending CN107741271A (en) | 2017-09-22 | 2017-09-22 | A kind of Winding in Power Transformer state evaluating method based on system delay Order- reduction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107741271A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108872777A (en) * | 2018-05-31 | 2018-11-23 | 浙江大学 | Winding in Power Transformer state evaluating method based on improved system delay Order- reduction |
| CN109991508A (en) * | 2019-04-15 | 2019-07-09 | 中国计量大学 | A kind of transformer winding state diagnostic method based on kinematic nonlinearity characteristic sequence |
| CN111273100A (en) * | 2020-02-20 | 2020-06-12 | 浙江大学 | Power transformer winding state evaluation method based on vibration phase |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103149476A (en) * | 2013-02-06 | 2013-06-12 | 浙江大学 | Electric-vibration model-based power transformer failure diagnosis method |
| CN104655967A (en) * | 2015-03-17 | 2015-05-27 | 国家电网公司 | Extraction method for vibration signal characteristic quantity of winding of distribution transformer |
-
2017
- 2017-09-22 CN CN201710867458.0A patent/CN107741271A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103149476A (en) * | 2013-02-06 | 2013-06-12 | 浙江大学 | Electric-vibration model-based power transformer failure diagnosis method |
| CN104655967A (en) * | 2015-03-17 | 2015-05-27 | 国家电网公司 | Extraction method for vibration signal characteristic quantity of winding of distribution transformer |
Non-Patent Citations (1)
| Title |
|---|
| 郑婧: ""基于盲分离的电力变压器振动模型研究"", 《中国优秀博士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108872777A (en) * | 2018-05-31 | 2018-11-23 | 浙江大学 | Winding in Power Transformer state evaluating method based on improved system delay Order- reduction |
| CN108872777B (en) * | 2018-05-31 | 2020-07-17 | 浙江大学 | Power transformer winding state evaluation method based on system delay order estimation |
| CN109991508A (en) * | 2019-04-15 | 2019-07-09 | 中国计量大学 | A kind of transformer winding state diagnostic method based on kinematic nonlinearity characteristic sequence |
| CN111273100A (en) * | 2020-02-20 | 2020-06-12 | 浙江大学 | Power transformer winding state evaluation method based on vibration phase |
| CN111273100B (en) * | 2020-02-20 | 2021-04-27 | 浙江大学 | Power transformer winding state evaluation method based on vibration phase |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9404957B2 (en) | Fault diagnosis and preliminary location system and method for transformer core looseness | |
| CN202735425U (en) | Power transformer fault detection system based on vibration | |
| CN100373166C (en) | Device for detecting transformer winding state utilizing sweep frequency power source exciting | |
| CN105259471B (en) | A kind of three-dimensional fault-line selecting method based on accidental resonance and transient current signal | |
| CN107741271A (en) | A kind of Winding in Power Transformer state evaluating method based on system delay Order- reduction | |
| CN102721464A (en) | Method and system for detecting winding deformation faults of power transformer | |
| KR100970943B1 (en) | System and method for recording fault of power system comprising generating fuel cell | |
| Zhang et al. | Operational modal analysis of transformer windings | |
| CN102721897A (en) | Diagnosis method and system of turn-to-turn short circuit fault of power transformer winding | |
| CN106338336A (en) | Transformer vibration on-line monitoring system | |
| Geng et al. | Modal parameters identification of power transformer winding based on improved empirical mode decomposition method | |
| CN106932184A (en) | A kind of Diagnosis Method of Transformer Faults based on improvement hierarchical clustering | |
| CN107883996A (en) | A kind of data acquisition and signal processing method and data acquisition unit | |
| Stein et al. | The torsional stress analyzer for continuously monitoring turbine-generators | |
| CN106443257A (en) | System for monitoring state of EHV (extra-high voltage) transformer based on vibration | |
| CN108872777A (en) | Winding in Power Transformer state evaluating method based on improved system delay Order- reduction | |
| CN110161351A (en) | Transformer winding fault pilot system and diagnostic method under a kind of oscillation wave | |
| CN111273100B (en) | Power transformer winding state evaluation method based on vibration phase | |
| Wu et al. | A Mechanical-Electromagnetic Coupling Model of Transformer Windings and Its application in the Vibration-based Condition Monitoring | |
| Jing et al. | Identification of the nonlinear vibration system of power transformers | |
| CN203025253U (en) | Capacitive equipment dielectric loss energized testing device | |
| Dhend et al. | Fault diagnosis methodology in smart grid with distributed energy generation | |
| Xie et al. | Online estimation of turbine–generator shaft fatigue loss-of-life caused by subsynchronous resonance | |
| CN105699838A (en) | A voltage transformer winding state detection method and an apparatus thereof | |
| Shao et al. | The fault detection of transformer windings' deformation based on Vibration frequency response analysis |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180227 |
|
| RJ01 | Rejection of invention patent application after publication |