CN114371426A - Transformer winding mechanical state detection method based on non-negative tensor decomposition - Google Patents
Transformer winding mechanical state detection method based on non-negative tensor decomposition Download PDFInfo
- Publication number
- CN114371426A CN114371426A CN202210026515.3A CN202210026515A CN114371426A CN 114371426 A CN114371426 A CN 114371426A CN 202210026515 A CN202210026515 A CN 202210026515A CN 114371426 A CN114371426 A CN 114371426A
- Authority
- CN
- China
- Prior art keywords
- calculating
- short
- tensor
- vibration signal
- domain envelope
- 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
- 238000004804 winding Methods 0.000 title claims abstract description 64
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 20
- 238000001514 detection method Methods 0.000 title description 10
- 238000000034 method Methods 0.000 claims abstract description 43
- 239000011159 matrix material Substances 0.000 claims description 25
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000011478 gradient descent method Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000000877 morphologic effect Effects 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 abstract description 5
- 230000015572 biosynthetic process Effects 0.000 abstract description 2
- 238000012360 testing method Methods 0.000 description 7
- 238000009413 insulation Methods 0.000 description 6
- 238000005070 sampling Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000009863 impact test Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000010339 dilation Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000011418 maintenance treatment Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Images
Classifications
-
- 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
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/62—Testing of transformers
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
Abstract
The application discloses a method for detecting the mechanical state of a transformer winding based on non-negative tensor decomposition, which comprises the following steps: and acquiring a short-circuit current signal and a vibration signal when the transformer winding is in short-circuit impact. And calculating to obtain the short-circuit current signal time domain envelope and the vibration signal time domain envelope according to the short-circuit current signal and the vibration signal. And constructing according to the short circuit current signal time domain envelope and the vibration signal time domain envelope to obtain a three-order non-negative tensor. And calculating the core tensor of the third-order non-negative tensor to obtain a first core tensor. And calculating the statistic of the first core tensor to obtain a first statistic. And calculating to obtain a first control limit according to the first statistic. And if the first statistic belongs to the interval of the first control limit, the mechanical state of the transformer winding is normal. According to the method and the device, the initial-stage fault of the transformer winding can be effectively identified, and then an effective operation and maintenance strategy is adopted, so that the formation of a major fault is avoided.
Description
Technical Field
The invention relates to the technical field of power equipment state detection, in particular to a method for detecting the mechanical state of a transformer winding based on non-negative tensor decomposition.
Background
Nowadays, large power transformers are the most important electrical devices in power grids, the performance of the large power transformers directly affects the safe and stable operation of power systems, and the reliability of the large power transformers is very important for long-term operation of the power transformers. With the continuous increase of the capacity of a power grid in China, the short-circuit capacity is correspondingly increased, and the huge electromagnetic acting force generated by the impact current formed by the short circuit at the outlet of the transformer can seriously threaten the mechanical strength and the stability of a transformer winding. If the fault transformer is not maintained in time, the transformer is damaged, the normal operation of a power grid is affected, and even the power system is broken down.
After the running transformer is subjected to multiple short circuit impacts, the winding of the running transformer is stressed and deformed, insulation defects are hidden, and insulation breakdown is possibly caused once voltage fluctuation occurs. Years of practical experience shows that after the transformer winding is deformed, the transformer winding is difficult to accurately judge through a general insulation test and an oil test, and the fault is represented as a latent fault. Therefore, in the operation process, when the transformer experiences an external short-circuit accident or is subjected to routine maintenance, how to effectively diagnose whether the transformer winding is loose or not and further judge whether the transformer needs to be maintained is an important measure for ensuring the safe operation of the transformer.
The deformation detection of the transformer winding is one of the conventional test items of the current transformer, and the most common detection methods mainly comprise two methods: the method is characterized in that a short-circuit impedance method is adopted, and whether the transformer winding changes to influence safe operation or not is judged through the change of impedance or leakage reactance by measuring and analyzing the short-circuit impedance or leakage reactance of the transformer winding under the power frequency voltage. And the second method is a frequency response analysis method, which is used for diagnosing whether the transformer winding changes or not by measuring the transfer function of the transformer winding and describing the transfer function from a frequency domain.
However, the above methods have problems mainly: firstly, the short-circuit impedance method has low sensitivity and low fault detection rate, and can only obtain accurate diagnosis results when the whole deformation of the transformer coil is serious. Secondly, the frequency response analysis method has complex frequency response waveform, needs more tests for judging the winding condition, and is difficult to form clear quantitative criterion, thereby being not beneficial to judgment of workers.
Disclosure of Invention
The invention provides a method for detecting the mechanical state of a transformer winding based on non-negative tensor decomposition, which aims to solve the problem of how to effectively diagnose whether the transformer winding is abnormal or not when the transformer experiences an external short circuit accident or is subjected to conventional maintenance in the operation process of the transformer. The invention effectively and accurately detects whether the mechanical state of the transformer winding is abnormal or not by diagnosing the mechanical working state of the transformer winding, thereby taking effective measures to the transformer winding in time and improving the operation reliability of the transformer winding.
The invention is realized by the following technical scheme:
the application provides a method for detecting the mechanical state of a transformer winding based on non-negative tensor decomposition, which comprises the following steps:
acquiring a short-circuit current signal and a vibration signal when a transformer winding is in short-circuit impact;
calculating to obtain a short-circuit current signal time domain envelope and a vibration signal time domain envelope according to the short-circuit current signal and the vibration signal;
constructing according to the short circuit current signal time domain envelope and the vibration signal time domain envelope to obtain a three-order non-negative tensor;
calculating a core tensor of the three-order non-negative tensor to obtain a first core tensor;
calculating statistics of the first core tensor to obtain a first statistic;
calculating to obtain a first control limit according to the first statistic;
judging the first statistic, if the first statistic belongs to the interval of the first control limit, the mechanical state of the transformer winding is normal;
and if the first statistic does not belong to the interval of the first control limit, the mechanical state of the transformer winding is abnormal.
Further, the calculation formula of the short-circuit current signal time domain envelope is as follows:
wherein i (t) is a short-circuit current signal ih(t) is the result of the Hilbert transform of the short circuit current signal.
Further, the method for calculating the time-domain envelope of the vibration signal comprises the following steps:
finding all local maximum value points in the vibration signal;
representing all positive local maxima and all negative local maxima of the local maxima points as a data sequence;
constructing a rectangular structural element according to the data sequence to obtain a first structural element;
performing multi-scale morphological wavelet decomposition on the vibration signal according to the first structural element to obtain a first scale signal;
calculating the vibration signal and the first scale signal to obtain a first correlation coefficient, and selecting a scale signal corresponding to the maximum value in the first correlation coefficient to obtain a second scale signal;
and calculating to obtain the time domain envelope of the vibration signal according to the second scale signal.
Further, the third-order non-negative tensor is constructed by the transpose of the time sequence of the vibration signal time domain envelope, and the short circuit current signal time domain envelope.
Further, the method for calculating the first core tensor comprises the following steps:
establishing a solution model of the first core tensor;
solving by using a gradient descent method to obtain a first objective function;
performing iterative computation on the first objective function, and if a convergence condition is met, outputting the first core tensor;
if the convergence condition is not met, the iterative computation is continued.
Further, the convergence condition is a preset iteration convergence threshold.
Further, the method for calculating the first statistic includes:
performing mode expansion matrix on the first core tensor to obtain a first matrix and a second matrix;
calculating an eigenvalue of the first matrix to obtain a first eigenvalue;
calculating the eigenvalue of the second matrix to obtain a second eigenvalue;
and constructing the first statistic according to the first characteristic value and the second characteristic value.
Further, the first control limit is that the first statistic is determined by applying a 3 σ criterion.
The vibration signal and the current signal under the transformer short-circuit impact are calculated and analyzed, Hilbert transform is adopted, the time domain envelope of the short-circuit current signal is extracted through amplitude demodulation, the time domain envelope of the vibration signal is calculated through adaptive non-sampling form size wave transform, the three-order nonnegative tensor is constructed by using the time domain envelope of the short-circuit current signal and the time domain envelope of the vibration signal, the core tensor of the three-order nonnegative tensor under sparse constraint is calculated based on a gradient descent method, the statistic and the control limit of the core tensor are constructed, and the mechanical state of the transformer winding is judged through the distribution of the control limit. The method has the advantages that the accurate detection of the mechanical state of the transformer is realized, so that the potential fault hazard of the transformer winding at the initial stage can be effectively identified, an effective operation and maintenance strategy is adopted, major faults are avoided, and the operation reliability of the transformer is improved.
Drawings
In order to more clearly explain the technical solution of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious to those skilled in the art that other drawings can be obtained according to the drawings without any creative effort.
Fig. 1 is a schematic flow chart of a method for detecting a mechanical state of a transformer winding based on non-negative tensor decomposition according to the present application;
FIG. 2 is a schematic diagram of vibration signals for detecting mechanical states of transformer windings provided herein;
FIG. 3 is a schematic diagram of the time-domain envelope of the vibration signal for detecting the mechanical state of the transformer winding provided by the present application;
FIG. 4 is a schematic flow chart of a method for calculating a time-domain envelope of a vibration signal according to the present application;
FIG. 5 is a flowchart illustrating a method for calculating a first core tensor according to the present application;
fig. 6 is a flowchart illustrating a method for calculating a first statistic according to the present application.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
After the running transformer is subjected to multiple short circuit impacts, the winding of the running transformer is stressed and deformed, insulation defects are hidden, and insulation breakdown is possibly caused once voltage fluctuation occurs. Years of practical experience shows that after the transformer winding is deformed, the transformer winding is difficult to accurately judge through a general insulation test and an oil test, and the fault is represented as a latent fault. Therefore, the method for detecting the mechanical state of the transformer winding based on the non-negative tensor decomposition can effectively and accurately detect whether the mechanical state of the transformer winding is abnormal or not by diagnosing the mechanical working state of the transformer winding, so that effective measures can be taken for the transformer winding in time, and the operation reliability of the transformer winding is improved. The technical scheme of the application is explained in detail as follows:
a certain 110kV transformer is used as a test object to carry out a short-circuit impact test, and a vibration signal and a current signal in the short-circuit impact test are tested, so that the detection method and the detection process of the mechanical state of the transformer winding are explained.
Referring to fig. 1, a schematic flow chart of a method for detecting a mechanical state of a transformer winding based on non-negative tensor decomposition according to an embodiment of the present application is shown;
as can be seen from fig. 1, the method for detecting the mechanical state of the transformer winding based on the non-negative tensor decomposition provided by the present application includes:
s1: acquiring a short-circuit current signal and a vibration signal when a transformer winding is in short-circuit impact; carrying out short circuit impact test on the transformer winding to obtain a short circuit current signal i (t) and a vibration signal s when the transformer winding is in short circuit impact0(t), as shown in fig. 2, provides the vibration signal for detecting the mechanical state of the transformer winding according to the embodiment of the present application. Wherein the short-circuit current signal has a length of N0Sampling frequency of f0Vibration signal length of N1Sampling frequency of f1(ii) a Here, f0＝10kHz，f1＝51200kHz，N0＝3000，N1＝30720。
S2: according to the short-circuit current signal i (t) and the vibration signal s0(t), calculating to obtain a short-circuit current signal time domain envelope and a vibration signal time domain envelope;
the calculation formula of the short-circuit current signal time domain envelope is as follows:
wherein i (t) is a short-circuit current signal ih(t) is the result of the Hilbert transform of the short circuit current signal.
As can be seen from fig. 3, a schematic diagram of the time-domain envelope of the vibration signal for detecting the mechanical state of the transformer winding is provided for the present application;
the embodiment of the application adopts the self-adaptive non-sampling form wavelet transform to calculate the time domain envelope of the vibration signal, and comprises the following specific steps:
referring to fig. 4, a schematic flow chart of a method for calculating a time-domain envelope of a vibration signal according to the present application is shown;
s21: finding all local maximum value points in the vibration signal; recording all local maximum value points as s1(M), where M is the number of all local maximum points in the vibration signal, and the method for confirming the local maximum points is as follows: obtaining a derivative of the reduced vibration signal to obtain a sequence sd(t) then calculating the product of two adjacent points in the sequence as csdi(t)，csdi(t)＝sdi(t)×sd(i-1)(t)，i＝1,2,L,N1-1, cs obtained after multiplicationdi(t) and the sequence sd(t) sequentially searching all local maximum points of the vibration signal in a positive and negative mode, wherein the specific method comprises the following steps:
when cs isdiIf (t) < 0, if csdi(t) < 0 and sd(i-1)(t) > 0, then s(i-1)(t) is a local maximum point;
when cs isdiWhen (t) > 0, s(i-1)(t) is a non-extreme point;
when cs isdiIf (t) is 0, then sd(i-1)(t) is 0, and two points s are calculatedi(t) and si-2(t) is the product of 0, let cs bedi(t)'＝si(t)×si-2(t) if csdi(t)' < 0 and sd(i-2)(t) > 0, then si-1(t) is the local maximum point, if sd(i-2)(t) is 0, then si-1And (t) is a non-extreme point.
S22: representing all positive local maxima and all negative local maxima of the local maxima points as a data sequence; the data sequences are respectively marked as ps (M)1) And ns (M)2) Here, M1 and M2 are the numbers of positive local maximums and negative local maximums, respectively.
S23: according to said data sequence ps (M)1) And ns (M)2) Constructing rectangular structural elements, and constructing rectangular structural elements with lengths and heights of L and H respectively to obtain a first structural element B;
the first structural element B is constructed by a matrix with the length L and the height H:
L＝2×ceil(0.5×min(L(i)))i＝1,2,L,min(M1,M2)；
H＝0.5×min(H(i))i＝1,2,L,min(M1,M2)；
L(i)＝ps(i)-ns(i)；
in the formula: abs represents the absolute value, ns (1) and ps (1) represent the first positive local maximum and the first negative local maximum, respectively, and h (i) represents the minimum after taking the absolute value when the first positive local maximum ns (1) is greater than the first negative local maximum ps (1). The other is to take the absolute minimum when the first positive local maximum ns (1) is less than the first negative local maximum ps (1).
S24: performing multi-scale morphological wavelet decomposition on the vibration signal according to the first structural element to obtain a first scale signal and a detail signal;
yj+1＝(id-T)(xj)；
in the formula: x is the number ofj+1And yj+1Respectively a scale signal and a detail signal on the j +1 th scale; id is an equivalence operator; t (-) represents the mathematical morphology operator, here taking the average of erosion and dilation; epsilon is a corrosion operator; δ is the dilation operator.
S25: calculating the vibration signal and the first scale signal to obtain a first correlation coefficient, and selecting a scale signal corresponding to the maximum value in the first correlation coefficient to obtain a second scale signal;
s26: and demodulating and calculating to obtain the time domain envelope of the vibration signal according to the second scale signal.
The vibration signal time domain envelope calculation formula is as follows:
in the formula: ys isRAnd ysIRespectively the real part and the imaginary part of the selected second scale signal.
And constructing according to the short circuit current signal time domain envelope and the vibration signal time domain envelope to obtain a three-order non-negative tensor. In particular, the third order non-negative tensor
Has a first step length of I3The transpose and the second order length of the time sequence of the time domain envelope of the vibration signal are I3And the third order length is I3And constructing the time domain envelope of the short-circuit current signal.
As can be seen from fig. 5, a flow chart of the method for calculating the first core tensor is illustrated in the present application;
s5: calculating the third order non-negative tensor
Obtaining a first core tensor;
s51: establishing a solution model of the first core tensor;
s.t.U(1)≥0,U(2)≥0,U(3)≥0；
in the formula:
in order to be the objective function, the target function,
a Frobenius norm representing a matrix; u shape(1)、U(2)And U(3)Projection tensors of 1 st, 2 nd and 3 rd orders, respectively; | | U(3)||1A 1 norm representing a 3 rd order projection tensor; lambda | U(3)||1And expressing sparse constraint terms for eliminating redundant terms in the 3 rd order projection tensor.
S52: solving by using a gradient descent method to obtain a first objective function;
s53: and performing iterative computation on the first objective function, and outputting the first core tensor if a convergence condition is met.
The initialization iteration number k is equal to 0, λ is equal to 0.01, and ∈ is equal to 0.001, where ∈ is a preset convergence threshold of falling and collection, which is also a convergence condition in this application;
randomly initializing the projection tensors of 1 st order, 2 nd order and 3 rd order, and respectively recording the projection tensors as
And
ensure non-negative;
let k equal to k +1, update the 1 st, 2 nd and 3 rd order projection tensors according to the following formula
And
is recorded as:
in the formula:
1 is a 3 × 3 all 1 matrix;
representing element-wise multiplication; e denotes dividing by element correspondence.
If the convergence condition is satisfied
Outputting the first core tensor and projection matrix U(1)、U(2)And U(3)Here, the first core tensor
Has dimension of C1×C2×C3；
S54: if the convergence condition is not met, the iterative computation is continued.
As can be seen from fig. 6, a flow chart of a method for calculating the first statistic is shown.
S6: calculating to obtain a first control limit according to the first statistic;
s61: performing mode expansion matrix on the first core tensor to obtain a first matrix and a second matrix;
combining the first core tensor
Performing modulo 1 and modulo 2 expansion to obtain dimension C1×C2C3And C2×C1C3Said first matrix G1And said second matrix G2
S62: calculating an eigenvalue of the first matrix to obtain a first eigenvalue;
calculating the first matrix G1Obtaining the first characteristic value lambda1,λ2,L,
Here, b1The number of the characteristic values;
s63: calculating the eigenvalue of the second matrix to obtain a second eigenvalue;
calculating the second matrix G2Obtaining the second characteristic value lambda1,λ2,L,
Here, b2The number of the characteristic values;
s64: constructing the first statistic Γ from the first and second eigenvalues;
in the formula: l is1Is composed of b1The number of zero elements in a diagonal matrix formed by the characteristic values; l is2Is composed of b2The number of zero elements in a diagonal matrix formed by the characteristic values; eta1And η2Is constant, here, η1＝η2＝0.5；
The first control limit is defined by the range of the first statistic's 3 σ criterion, which is a common mathematical method and will not be described in detail here. Then, judging whether the first statistic belongs to the interval range of the first control limit, and if the first statistic belongs to the interval range of the first control limit, indicating that the mechanical state of the transformer winding is normal; if not, the mechanical state of the transformer winding is abnormal and changes, and the maintenance treatment needs to be carried out in time at the moment, so that the formation of major faults is avoided.
The vibration signal and the current signal under the transformer short-circuit impact are calculated and analyzed, Hilbert transform is adopted, the time domain envelope of the short-circuit current signal is extracted through amplitude demodulation, the time domain envelope of the vibration signal is calculated through adaptive non-sampling form size wave transform, the three-order nonnegative tensor is constructed by using the time domain envelope of the short-circuit current signal and the time domain envelope of the vibration signal, the core tensor of the three-order nonnegative tensor under sparse constraint is calculated based on a gradient descent method, the statistic and the control limit of the core tensor are constructed, and the mechanical state of the transformer winding is judged through the distribution of the control limit. The method has the advantages that the accurate detection of the mechanical state of the transformer winding is realized, so that the potential fault hazard of the transformer winding at the initial stage can be effectively identified, an effective operation and maintenance strategy is adopted, major faults are avoided, and the operation reliability of the transformer is improved.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is limited only by the appended claims.
Claims (8)
1. A method for detecting a mechanical state of a transformer winding based on non-negative tensor decomposition, the method comprising:
acquiring a short-circuit current signal and a vibration signal when a transformer winding is in short-circuit impact;
calculating to obtain a short-circuit current signal time domain envelope and a vibration signal time domain envelope according to the short-circuit current signal and the vibration signal;
constructing according to the short circuit current signal time domain envelope and the vibration signal time domain envelope to obtain a three-order non-negative tensor;
calculating a core tensor of the three-order non-negative tensor to obtain a first core tensor;
calculating statistics of the first core tensor to obtain a first statistic;
calculating to obtain a first control limit according to the first statistic;
if the first statistic belongs to the interval of the first control limit, the mechanical state of the transformer winding is normal;
and if the first statistic does not belong to the interval of the first control limit, the mechanical state of the transformer winding is abnormal.
2. The method for detecting the mechanical state of the winding of the transformer based on the non-negative tensor decomposition as recited in claim 1, wherein a calculation formula of the time-domain envelope of the short-circuit current signal is as follows:
wherein i (t) is a short-circuit current signal ih(t) is the result of the Hilbert transform of the short circuit current signal.
3. The method for detecting the mechanical state of the winding of the transformer based on the non-negative tensor decomposition as recited in claim 1, wherein the method for calculating the time-domain envelope of the vibration signal comprises the following steps:
acquiring all local maximum value points in the vibration signal;
representing all positive local maxima and all negative local maxima of the local maxima points as a data sequence;
constructing a rectangular structural element according to the data sequence to obtain a first structural element;
performing multi-scale morphological wavelet decomposition on the vibration signal according to the first structural element to obtain a first scale signal;
calculating the vibration signal and the first scale signal to obtain a first correlation coefficient, and selecting a scale signal corresponding to the maximum value in the first correlation coefficient to obtain a second scale signal;
and calculating to obtain the time domain envelope of the vibration signal according to the second scale signal.
4. The method for detecting the mechanical state of the winding of the transformer based on the non-negative tensor decomposition as recited in claim 1, wherein the third-order non-negative tensor is constructed by transposing a time sequence of the time-domain envelope of the vibration signal, constructing the time sequence of the time-domain envelope of the vibration signal, and constructing the time-domain envelope of the short-circuit current signal.
5. The method for detecting the mechanical state of the winding of the transformer based on the non-negative tensor decomposition as recited in claim 1, wherein the calculating method of the first core tensor comprises the following steps:
establishing a solution model of the first core tensor;
solving by using a gradient descent method to obtain a first objective function;
performing iterative computation on the first objective function, and if a convergence condition is met, outputting the first core tensor;
if the convergence condition is not met, the iterative computation is continued.
6. The method for detecting the mechanical state of the winding of the transformer based on the non-negative tensor decomposition as recited in claim 5, wherein the convergence condition is a preset iterative convergence threshold.
7. The method for detecting the mechanical state of the winding of the transformer based on the non-negative tensor decomposition as recited in claim 1, wherein the calculating method of the first statistic comprises the following steps:
performing mode expansion matrix on the first core tensor to obtain a first matrix and a second matrix;
calculating an eigenvalue of the first matrix to obtain a first eigenvalue;
calculating the eigenvalue of the second matrix to obtain a second eigenvalue;
and constructing the first statistic according to the first characteristic value and the second characteristic value.
8. The method of claim 1, wherein the first control limit is determined by applying a 3 σ criterion to the first statistic.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210026515.3A CN114371426A (en) | 2022-01-11 | 2022-01-11 | Transformer winding mechanical state detection method based on non-negative tensor decomposition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210026515.3A CN114371426A (en) | 2022-01-11 | 2022-01-11 | Transformer winding mechanical state detection method based on non-negative tensor decomposition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114371426A true CN114371426A (en) | 2022-04-19 |
Family
ID=81144399
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210026515.3A Pending CN114371426A (en) | 2022-01-11 | 2022-01-11 | Transformer winding mechanical state detection method based on non-negative tensor decomposition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114371426A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114814659A (en) * | 2022-05-25 | 2022-07-29 | 云南电网有限责任公司电力科学研究院 | Method and device for monitoring state of transformer winding under short circuit impact |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010060253A1 (en) * | 2008-11-25 | 2010-06-03 | 上海市电力公司 | The system and method for detecting the state of the transformer winding by utilizing excitation of the constant current sweep frequency power source |
| US20100185578A1 (en) * | 2009-01-22 | 2010-07-22 | Nec Laboratories America, Inc. | Social network analysis with prior knowledge and non-negative tensor factorization |
| CN105182172A (en) * | 2015-07-23 | 2015-12-23 | 广东电网有限责任公司电力科学研究院 | Vibration signal pattern spectrum-based method for diagnosing winding conditions under sudden short circuit of transformer |
| CN110824383A (en) * | 2019-11-07 | 2020-02-21 | 国网天津市电力公司 | Method for calculating short circuit impact cumulative effect of transformer |
| US20200200813A1 (en) * | 2018-12-21 | 2020-06-25 | Zhejiang University | Online diagnosis method for deformation position on trasnformation winding |
| CN111415039A (en) * | 2020-03-19 | 2020-07-14 | 北京航空航天大学 | Flight delay mode analysis method based on non-negative tensor decomposition |
-
2022
- 2022-01-11 CN CN202210026515.3A patent/CN114371426A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010060253A1 (en) * | 2008-11-25 | 2010-06-03 | 上海市电力公司 | The system and method for detecting the state of the transformer winding by utilizing excitation of the constant current sweep frequency power source |
| US20100185578A1 (en) * | 2009-01-22 | 2010-07-22 | Nec Laboratories America, Inc. | Social network analysis with prior knowledge and non-negative tensor factorization |
| CN105182172A (en) * | 2015-07-23 | 2015-12-23 | 广东电网有限责任公司电力科学研究院 | Vibration signal pattern spectrum-based method for diagnosing winding conditions under sudden short circuit of transformer |
| US20200200813A1 (en) * | 2018-12-21 | 2020-06-25 | Zhejiang University | Online diagnosis method for deformation position on trasnformation winding |
| CN110824383A (en) * | 2019-11-07 | 2020-02-21 | 国网天津市电力公司 | Method for calculating short circuit impact cumulative effect of transformer |
| CN111415039A (en) * | 2020-03-19 | 2020-07-14 | 北京航空航天大学 | Flight delay mode analysis method based on non-negative tensor decomposition |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114814659A (en) * | 2022-05-25 | 2022-07-29 | 云南电网有限责任公司电力科学研究院 | Method and device for monitoring state of transformer winding under short circuit impact |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106443316B (en) | Multi-information detection method and device for deformation state of power transformer winding | |
| CN112201260B (en) | Transformer running state online detection method based on voiceprint recognition | |
| CN103645249B (en) | Online fault detection method for reduced set-based downsampling unbalance SVM (Support Vector Machine) transformer | |
| CN109507554B (en) | Electrical equipment insulation state evaluation method | |
| CN106529124A (en) | A transformer insulation state assessment method based on principal component analysis and support vector machines | |
| CN111679158A (en) | Power distribution network fault identification method based on synchronous measurement data similarity | |
| CN107370150A (en) | The Power system state estimation Bad data processing method measured based on synchronized phasor | |
| CN112147444B (en) | Power transformer working state monitoring method and system | |
| CN106548021A (en) | A kind of electronic mutual inductor error state appraisal procedure based on higher-dimension random matrix | |
| CN112183590A (en) | Transformer fault diagnosis method based on Oneclass SVM algorithm | |
| CN113138322A (en) | Method, device, equipment and storage medium for detecting high-resistance earth fault of power distribution network | |
| CN112731260A (en) | Online evaluation method for error state of voltage transformer based on concept drift recognition | |
| CN112149549A (en) | GIS partial discharge type identification method based on depth residual error network | |
| CN113391244A (en) | VMD-based transformer switching-on vibration signal characteristic frequency calculation method | |
| CN114371426A (en) | Transformer winding mechanical state detection method based on non-negative tensor decomposition | |
| CN116643091A (en) | Online insulation monitoring method for converter transformer valve side sleeve based on multidimensional time sequence characteristics | |
| CN117748507A (en) | Distribution network harmonic access uncertainty assessment method based on Gaussian regression model | |
| CN110161351B (en) | Transformer winding fault test system and diagnosis method under oscillatory wave | |
| CN114609483A (en) | Hilbert transform-based GIS partial discharge signal feature extraction method | |
| CN110703149A (en) | Method and system for detecting vibration and sound of running state of transformer by utilizing character spacing | |
| CN114384319A (en) | Grid-connected inverter island detection method, system, terminal and medium | |
| CN112034312A (en) | Power equipment insulation defect mode identification method | |
| Abidullah et al. | Real-time power quality disturbances detection and classification system | |
| CN110017894B (en) | Method and device for filtering random noise in vibration and sound detection of transformer in running state | |
| CN116910470A (en) | GIS combined electrical apparatus partial discharge fault mode identification method |
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 |