CN114137444B - Transformer running state monitoring method and system based on acoustic signals - Google Patents
Transformer running state monitoring method and system based on acoustic signals Download PDFInfo
- Publication number
- CN114137444B CN114137444B CN202111434964.3A CN202111434964A CN114137444B CN 114137444 B CN114137444 B CN 114137444B CN 202111434964 A CN202111434964 A CN 202111434964A CN 114137444 B CN114137444 B CN 114137444B
- Authority
- CN
- China
- Prior art keywords
- acoustic signal
- filter
- transformer
- characteristic coefficient
- covariance matrix
- 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 34
- 238000012544 monitoring process Methods 0.000 title claims abstract description 26
- 239000011159 matrix material Substances 0.000 claims abstract description 140
- 238000001228 spectrum Methods 0.000 claims abstract description 35
- 239000013598 vector Substances 0.000 claims description 40
- 230000005236 sound signal Effects 0.000 claims description 26
- 238000005316 response function Methods 0.000 claims description 22
- 238000000354 decomposition reaction Methods 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000017105 transposition Effects 0.000 claims description 6
- 238000004804 winding Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 210000005069 ears Anatomy 0.000 description 2
- 239000003921 oil Substances 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 238000010183 spectrum analysis Methods 0.000 description 2
- 229910000976 Electrical steel Inorganic materials 0.000 description 1
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005520 electrodynamics Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000011418 maintenance treatment Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
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
-
- 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)
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention relates to a transformer running state monitoring method and system based on acoustic signals, which belong to the technical field of transformers, wherein a filter bank is designed by collecting the acoustic signals of the running transformer, characteristic coefficients in the frequency spectrum range of the acoustic signals covered by each filter in the filter bank are calculated to form an acoustic signal characteristic coefficient matrix, so that a covariance matrix of the acoustic signal characteristic coefficient matrix and a principal component information module value of the covariance matrix are determined, and finally, the running state of the transformer is judged according to the principal component information module value. The invention efficiently and accurately monitors the running state of the transformer by monitoring the acoustic signal of the running transformer in real time, thereby being capable of overhauling or replacing the transformer in time and improving the running reliability of the transformer.
Description
Technical Field
The invention relates to the technical field of transformers, in particular to a transformer running state monitoring method and system based on acoustic signals.
Background
Transformers are one of the most important devices in an electrical power system, and the stability of their operation has a significant impact on the safety of the electrical power system. The state monitoring is carried out on the transformer, the running state of the transformer can be mastered in real time, and the fault early warning is timely carried out, so that the fault early warning is prevented. Meanwhile, the operation maintenance and the state overhaul of the transformer are guided, the unplanned power failure is avoided, the service life of the transformer is prolonged, and the method has great significance in ensuring the safe and stable operation of the transformer and a power system.
In the running process of the transformer, winding vibration caused by electrodynamic force and periodic vibration caused by magnetostriction of the iron core silicon steel sheet are transmitted to the wall of a transformer oil tank through a transformer structural member and insulating oil, and are radiated to the periphery through air together with mechanical vibration and the like caused by a transformer cooling system to form an acoustic signal. Wherein, 20 Hz-20 kHz is audible frequency range of human ears, and the experienced transformer substation staff can directly listen to the sound of the running transformer by the ears to judge whether the state is normal or not. The audible sound signals can be conveniently obtained through the microphone sensors arranged on the periphery of the transformer, and the audible sound signals are not in direct contact with the transformer, so that the audible sound signals have the characteristics of flexible and convenient field implementation, no interference to the normal operation of the transformer, high timeliness and the like, and are increasingly brought into focus in the field of transformer state monitoring. However, due to the complexity of the mechanical structure of the transformer, the dispersion of the process, the diversity of the operating environment of the transformer station, and other factors, how to analyze and determine the operating state of the operating transformer from the acoustic signals of the operating transformer has been a research difficulty.
Disclosure of Invention
The invention aims to provide a method and a system for monitoring the running state of a transformer based on acoustic signals, so that the running state of the transformer can be judged efficiently and accurately by monitoring the acoustic signals of the running transformer in real time.
In order to achieve the above object, the present invention provides the following solutions:
a method of monitoring an operational state of a transformer based on acoustic signals, the method comprising:
collecting an acoustic signal of the transformer in operation by using a microphone sensor;
determining a filter bank from the acoustic signal;
taking the acoustic signals as input of a filter bank, calculating characteristic coefficients in a frequency spectrum range of the acoustic signals covered by each filter in the filter bank, and forming an acoustic signal characteristic coefficient matrix;
determining a covariance matrix of the acoustic signal characteristic coefficient matrix, and calculating a principal component information modulus value of the covariance matrix;
if the principal component information module value is larger than or equal to a preset module value threshold value, judging that the running state of the transformer is normal;
and if the principal component information module value is smaller than a preset module value threshold value, judging that the running state of the transformer is abnormal.
Optionally, determining a filter bank according to the acoustic signal specifically includes:
performing discrete Fourier transform on the acoustic signal to obtain spectrum distribution of the acoustic signal;
according to the spectrum distribution of the acoustic signal, determining a discrete time response function and a center frequency of each filter in the filter bank as follows:
wherein,as a discrete time response function of the jth filter, n is time, a j A is the scale factor of the jth filter j ＝f L /f cj ，f L Is the lowest center frequency of the filter bank, f cj For the center frequency of the jth filter, b is a time shift factor, alpha and beta are respectively a first positive real number and a second positive real number, theta is an initial phase, u () is a unit step function, Q 0 As a quality factor, B 0 For minimum bandwidth, P is the number of filters.
Optionally, taking the acoustic signal as an input of a filter bank, calculating characteristic coefficients in a frequency spectrum range of the acoustic signal covered by each filter in the filter bank to form an acoustic signal characteristic coefficient matrix, and specifically including:
taking the acoustic signal as input of a filter bank and using the formulaCalculating characteristic coefficients in the frequency spectrum range of the acoustic signal covered by each filter in the filter bank to form a characteristic coefficient vector of each filter; wherein F is i (b) Is the characteristic coefficient vector of the ith filter, s (k) is the acoustic signal at the kth moment, N is the acoustic signal length,/for the filter>A discrete time response function corresponding to the acoustic signal at the kth moment of the ith filter;
the eigenvector of each filter is used as a row vector, and all the row vectors form an acoustic signal eigenvalue matrix.
Optionally, determining a covariance matrix of the acoustic signal characteristic coefficient matrix, and calculating a principal component information modulus value of the covariance matrix, which specifically includes:
determining the covariance matrix of the characteristic coefficient matrix of the acoustic signal asWherein R is covariance matrix, N is acoustic signal length, F is acoustic signal characteristic coefficient matrix, and T represents matrix transposition;
performing eigenvalue decomposition on the covariance matrix to obtain a plurality of eigenvalues and eigenvectors of the covariance matrix;
based on the eigenvalues and eigenvectors of the covariance matrix, the formula is usedCalculating a principal component information module value of the covariance matrix; wherein ζ is the principal component information module value, λ 1 And lambda (lambda) W Respectively the maximum and minimum eigenvalues in the eigenvalues, |z 1 I represents the feature vector z 1 Is a mold of (a).
A transformer operating condition monitoring system based on acoustic signals, the system comprising:
the sound signal acquisition module is used for acquiring sound signals of the transformer in operation by using the microphone sensor;
a filter bank determination module for determining a filter bank from the acoustic signal;
the sound signal characteristic coefficient matrix forming module is used for taking the sound signal as the input of the filter bank, calculating characteristic coefficients in the frequency spectrum range of the sound signal covered by each filter in the filter bank, and forming a sound signal characteristic coefficient matrix;
the principal component information module value calculating module is used for determining a covariance matrix of the acoustic signal characteristic coefficient matrix and calculating a principal component information module value of the covariance matrix;
the normal operation judging module is used for judging that the operation state of the transformer is normal if the principal component information module value is larger than or equal to a preset module value threshold value;
and the abnormal operation judging module is used for judging that the operation state of the transformer is abnormal if the principal element information module value is smaller than a preset module value threshold value.
Optionally, the filter bank determining module specifically includes:
the frequency spectrum distribution obtaining submodule is used for carrying out discrete Fourier transform on the acoustic signals to obtain frequency spectrum distribution of the acoustic signals;
the discrete time response function determining submodule is used for determining that the discrete time response function and the center frequency of each filter in the filter bank are respectively as follows:
wherein,as a discrete time response function of the jth filter, n is time, a j A is the scale factor of the jth filter j ＝f L /f cj ，f L Is the lowest center frequency of the filter bank, f cj For the center frequency of the jth filter, b is a time shift factor, alpha and beta are respectively a first positive real number and a second positive real number, theta is an initial phase, u () is a unit step function, Q 0 As a quality factor, B 0 For minimum bandwidth, P is the number of filters.
Optionally, the acoustic signal characteristic coefficient matrix forming module specifically includes:
the characteristic coefficient vectors form sub-modules for taking the acoustic signals as the input of the filter bank and using the formulaCalculating characteristic coefficients in the frequency spectrum range of the acoustic signal covered by each filter in the filter bank to form a characteristic coefficient vector of each filter; wherein F is i (b) Is the characteristic coefficient vector of the ith filter, s (k) is the acoustic signal at the kth moment, N is the acoustic signal length,/for the filter>A discrete time response function corresponding to the acoustic signal at the kth moment of the ith filter;
the acoustic signal characteristic coefficient matrix forms a sub-module, and the characteristic coefficient vector for each filter is used as a row vector, and all row vectors form the acoustic signal characteristic coefficient matrix.
Optionally, the principal component information module calculating module specifically includes:
a covariance matrix determination submodule for determining a covariance matrix of the acoustic signal characteristic coefficient matrix asWherein R is covariance matrix, N is acoustic signal length, F is acoustic signal characteristic coefficient matrix, and T represents matrix transposition;
the eigenvalue decomposition sub-module is used for carrying out eigenvalue decomposition on the covariance matrix to obtain a plurality of eigenvalues and eigenvectors of the covariance matrix;
a principal component information modulus value calculation sub-module for utilizing a formula according to a plurality of eigenvalues and eigenvectors of the covariance matrixCalculating a principal component information module value of the covariance matrix; wherein ζ is the principal component information module value, λ 1 And lambda (lambda) W Respectively the maximum and minimum eigenvalues in the eigenvalues, |z 1 I represents the feature vector z 1 Is a mold of (a).
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention discloses a transformer running state monitoring method and system based on acoustic signals, which are characterized in that a filter bank is designed by collecting acoustic signals of a running transformer, characteristic coefficients in a frequency spectrum range of acoustic signals covered by each filter in the filter bank are calculated to form an acoustic signal characteristic coefficient matrix, a covariance matrix of the acoustic signal characteristic coefficient matrix and a principal component information module value of the covariance matrix are further determined, and finally, the running state of the transformer is judged according to the principal component information module value. The invention efficiently and accurately monitors the running state of the transformer by monitoring the acoustic signal of the running transformer in real time, thereby being capable of overhauling or replacing the transformer in time and improving the running reliability of the transformer.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for monitoring the operation state of a transformer based on acoustic signals;
FIG. 2 is a schematic diagram of a method for monitoring the operation state of a transformer based on acoustic signals according to an embodiment of the present invention;
FIG. 3 is a time domain diagram of an acoustic signal provided by an embodiment of the present invention;
fig. 4 is a spectrum diagram of an acoustic signal according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to provide a method and a system for monitoring the running state of a transformer based on acoustic signals, so that the running state of the transformer can be judged efficiently and accurately by monitoring the acoustic signals of the running transformer in real time.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
The invention provides a transformer running state monitoring method based on acoustic signals, as shown in fig. 1, the method comprises the following steps:
step 101, collecting an acoustic signal of a transformer in operation by using a microphone sensor.
Step 102, a filter bank is determined from the acoustic signal.
The method specifically comprises the following steps:
performing discrete Fourier transform on the acoustic signal to obtain the spectrum distribution of the acoustic signal;
according to the frequency spectrum distribution of the acoustic signal, determining a discrete time response function and a center frequency of each filter in the filter bank as follows:
wherein,as a discrete time response function of the jth filter, n is time, a j A is the scale factor of the jth filter j ＝f L /f cj ，f L Is the lowest center frequency of the filter bank, f cj For the center frequency of the jth filter, b is a time shift factor, alpha and beta are respectively a first positive real number and a second positive real number, theta is an initial phase, u () is a unit step function, Q 0 As a quality factor, B 0 For minimum bandwidth, P is the number of filters.
Step 103, taking the acoustic signals as the input of the filter bank, calculating characteristic coefficients in the frequency spectrum range of the acoustic signals covered by each filter in the filter bank, and forming an acoustic signal characteristic coefficient matrix.
The method specifically comprises the following steps:
taking the acoustic signal as input to the filter bank and using the formulaCalculating the spectral range of the acoustic signal covered by each filter in the filter bankThe characteristic coefficients in the filter constitute characteristic coefficient vectors of each filter; wherein F is i (b) Is the characteristic coefficient vector of the ith filter, s (k) is the acoustic signal at the kth moment, N is the acoustic signal length,/for the filter>A discrete time response function corresponding to the acoustic signal at the kth moment of the ith filter;
the eigenvector of each filter is used as a row vector, and all the row vectors form an acoustic signal eigenvalue matrix.
Step 104, determining a covariance matrix of the acoustic signal characteristic coefficient matrix, and calculating a principal component information modulus value of the covariance matrix.
The method specifically comprises the following steps:
determining covariance matrix of characteristic coefficient matrix of acoustic signal asWherein R is covariance matrix, N is acoustic signal length, F is acoustic signal characteristic coefficient matrix, and T represents matrix transposition;
performing eigenvalue decomposition on the covariance matrix to obtain a plurality of eigenvalues and eigenvectors of the covariance matrix;
based on the eigenvalues and eigenvectors of the covariance matrix, the formula is usedCalculating a principal component information module value of the covariance matrix; wherein ζ is the principal component information module value, λ 1 And lambda (lambda) W Respectively the maximum and minimum eigenvalues in the eigenvalues, |z 1 I represents the feature vector z 1 Is a mold of (a).
And 105, judging that the operation state of the transformer is normal if the principal component information module value is greater than or equal to a preset module value threshold value.
And 106, if the principal component information module value is smaller than a preset module value threshold value, judging that the running state of the transformer is abnormal.
According to the invention, the frequency spectrum analysis is carried out on the acoustic signals of the transformer in a certain time period, a filter bank is designed according to the frequency spectrum analysis, a covariance matrix is constructed according to the characteristic coefficient matrix of the acoustic signals after passing through the filter bank, and the working state of the transformer can be judged by calculating the change of the principal component information modulus of the covariance matrix constructed by the characteristic coefficient matrix of the acoustic signals.
The method for judging the working state of the transformer directly by monitoring the acoustic signals of the transformer in real time is efficient and accurate, is easy to implement, and is convenient for operators to find the abnormality of the transformer winding in time, so that the transformer is overhauled in time according to the abnormality, and the failure damage rate of the transformer is greatly reduced.
In order to further explain the invention in detail, the 110kV transformer of a transformer substation of a certain power company is used as an object to carry out on-line monitoring, and as shown in fig. 2, the working state of the transformer is judged according to the following steps:
(1) The microphone sensor is used for collecting the transformer acoustic signal s (N) in operation according to national standard GB/T1094.101-2008, and the sampling frequency is f s The length of the sound signal is N; the microphone sensor is connected to a signal acquisition system through a cable to condition, anti-aliasing filter and acquire the sound signals, the signal acquisition system is connected to a signal analysis terminal through a cable to store and display the acquired transformer sound signals, as shown in fig. 2; here, f s ＝51200Hz，N＝102400；
After the acoustic signal is acquired, the acoustic signal is required to be standardized, and the standardized formula is that
In the method, in the process of the invention,is the average value of the acoustic signal x (t); y (t) is the normalized acoustic signal.
(2) Designing a filter bank according to the frequency spectrum distribution of the acoustic signals, and calculating a characteristic coefficient matrix of the acoustic signals after passing through the filter bank, wherein the calculation process of the characteristic coefficient matrix of the acoustic signals is as follows:
performing discrete Fourier transform on the acoustic signals to obtain the frequency spectrum distribution of the transformer acoustic signals, as shown in fig. 2; the discrete fourier transform described in this step is a mathematical method commonly used in the art, and thus the inventors do not describe in detail here;
designing a filter bank based on the spectral distribution of the acoustic signal, wherein the discrete time response function of the jth filterCorresponding center frequency f cj The method comprises the following steps of:
1≤j≤P
wherein: f (f) L And f H The lowest center frequency and the cut-off frequency of the filter bank are determined by the frequency spectrum distribution of the transformer acoustic signal; b is a time shift factor ranging from 1 to N; the filter bank is preferably a cochlear filter bank.
2c, taking the acoustic signals as input of a filter bank, calculating characteristic coefficients in a frequency spectrum range covered by each filter, and obtaining an acoustic signal characteristic coefficient matrix F P×N The number of rows of the acoustic signal characteristic coefficient matrix is P, the number of columns is N (the number of columns of the acoustic signal characteristic coefficient matrix is the same as the length of the acoustic signal), and the kth row vector can be expressed as
(3) Calculating covariance matrix R of acoustic signal characteristic coefficient matrix P×P The number of rows and columns of the covariance matrix are P, and the calculation is performedThe formula is
Where T represents the matrix transpose.
(4) Performing eigenvalue decomposition on covariance matrix of acoustic signal eigenvalue to obtain W eigenvalues lambda of covariance matrix 1 ,λ 2 ,…,λ W And feature vector z 1 ,z 2 ,…,z W The W eigenvalues of the covariance matrix meet lambda 1 ＞λ 2 ＞…＞λ W ；
(5) The principal element information modulus value zeta of the covariance matrix of the characteristic parameters of the acoustic signal is calculated, and the calculation formula is as follows
In the formula, |z 1 I represents the feature vector z 1 Is a mold of (a).
(6) Judging the mechanical state of the transformer winding according to the change of the principal component information modulus value of the covariance matrix of the acoustic signal characteristic parameters of the transformer: if the principal element information modulus zeta is larger than 0.8, judging that the working state of the transformer is normal; if the primary element information module value zeta is smaller than 0.8, the working state of the transformer is judged to be changed, and at the moment, maintenance treatment is needed in time, so that major faults are avoided.
Here, the calculated principal component information modulus value ζ=0.89 of the covariance matrix of the acoustic signal characteristic parameters of the transformer indicates that the working state of the transformer is normal.
The invention discloses a method and a system for monitoring an acoustic signal of the running state of a transformer, which comprise the following steps: collecting the acoustic signals of the running transformer; carrying out standardization processing on the vibration signals; designing a filter bank according to the frequency spectrum distribution of the acoustic signals, and calculating a characteristic coefficient matrix of the acoustic signals after passing through the filter bank; performing eigenvalue decomposition on a covariance matrix of the acoustic signal characteristic parameter matrix to obtain an eigenvalue of the covariance matrix and a covariance matrix of the acoustic signal characteristic coefficient matrix calculated by the eigenvector; calculating principal element information modulus values of the covariance matrix of the characteristic parameters of the acoustic signals; and judging the mechanical state of the transformer winding according to the change of the principal component information modulus value of the covariance matrix of the acoustic signal characteristic parameters of the transformer. The method can effectively monitor the running state of the transformer on line with high sensitivity, thereby being capable of overhauling or replacing the transformer in time and improving the running reliability of the transformer.
The invention also provides a transformer running state monitoring system based on the acoustic signal, which comprises:
the sound signal acquisition module is used for acquiring sound signals of the transformer in operation by using the microphone sensor;
a filter bank determining module for determining a filter bank from the acoustic signal;
the sound signal characteristic coefficient matrix forming module is used for taking sound signals as input of the filter bank, calculating characteristic coefficients in the frequency spectrum range of the sound signals covered by each filter in the filter bank, and forming the sound signal characteristic coefficient matrix;
the principal component information module value calculating module is used for determining a covariance matrix of the acoustic signal characteristic coefficient matrix and calculating a principal component information module value of the covariance matrix;
the normal operation judging module is used for judging that the operation state of the transformer is normal if the principal element information module value is larger than or equal to a preset module value threshold value;
and the abnormal operation judging module is used for judging that the operation state of the transformer is abnormal if the principal element information module value is smaller than a preset module value threshold value.
The filter bank determining module specifically comprises:
the frequency spectrum distribution obtaining submodule is used for carrying out discrete Fourier transform on the acoustic signals to obtain frequency spectrum distribution of the acoustic signals;
the discrete time response function determining submodule is used for determining that the discrete time response function and the center frequency of each filter in the filter bank are respectively as follows:
wherein,as a discrete time response function of the jth filter, n is time, a j A is the scale factor of the jth filter j ＝f L /f cj ，f L Is the lowest center frequency of the filter bank, f cj For the center frequency of the jth filter, b is a time shift factor, alpha and beta are respectively a first positive real number and a second positive real number, theta is an initial phase, u () is a unit step function, Q 0 As a quality factor, B 0 For minimum bandwidth, P is the number of filters.
The acoustic signal characteristic coefficient matrix forming module specifically comprises:
the characteristic coefficient vectors form sub-modules for taking the acoustic signals as the input of the filter bank and using the formulaCalculating characteristic coefficients in the frequency spectrum range of the acoustic signal covered by each filter in the filter bank to form characteristic coefficient vectors of each filter; wherein F is i (b) Is the characteristic coefficient vector of the ith filter, s (k) is the acoustic signal at the kth moment, N is the acoustic signal length,/for the filter>A discrete time response function corresponding to the acoustic signal at the kth moment of the ith filter;
the acoustic signal characteristic coefficient matrix forms a sub-module, and the characteristic coefficient vector for each filter is used as a row vector, and all row vectors form the acoustic signal characteristic coefficient matrix.
The principal component information module value calculating module specifically comprises:
a covariance matrix determination submodule for determining a covariance matrix of the acoustic signal characteristic coefficient matrix asWherein R is a covariance matrix, N is the column number of the covariance matrix, F is an acoustic signal characteristic coefficient matrix, and T represents matrix transposition;
the eigenvalue decomposition sub-module is used for carrying out eigenvalue decomposition on the covariance matrix to obtain a plurality of eigenvalues and eigenvectors of the covariance matrix;
a principal component information modulus value calculation sub-module for utilizing a formula according to a plurality of eigenvalues and eigenvectors of the covariance matrixCalculating a principal component information module value of the covariance matrix; wherein ζ is the principal component information module value, λ 1 And lambda (lambda) W Respectively the maximum and minimum eigenvalues in the eigenvalues, |z 1 I represents the feature vector z 1 Is a mold of (a).
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.
Claims (4)
1. A method for monitoring the operational state of a transformer based on acoustic signals, the method comprising:
collecting an acoustic signal of the transformer in operation by using a microphone sensor;
determining a filter bank from the acoustic signal;
taking the acoustic signals as input of a filter bank, calculating characteristic coefficients in a frequency spectrum range of the acoustic signals covered by each filter in the filter bank, and forming an acoustic signal characteristic coefficient matrix;
determining a covariance matrix of the acoustic signal characteristic coefficient matrix, and calculating a principal component information modulus value of the covariance matrix;
if the principal component information module value is larger than or equal to a preset module value threshold value, judging that the running state of the transformer is normal;
if the principal component information module value is smaller than a preset module value threshold value, judging that the running state of the transformer is abnormal;
taking the acoustic signals as input of a filter bank, calculating characteristic coefficients in a frequency spectrum range of the acoustic signals covered by each filter in the filter bank to form an acoustic signal characteristic coefficient matrix, wherein the method specifically comprises the following steps:
taking the acoustic signal as input of a filter bank and using the formulaCalculating characteristic coefficients in the frequency spectrum range of the acoustic signal covered by each filter in the filter bank to form a characteristic coefficient vector of each filter; wherein F is i (b) Is the characteristic coefficient vector of the ith filter, s (k) is the acoustic signal at the kth moment, N is the acoustic signal length,/for the filter>A discrete time response function corresponding to the acoustic signal at the kth moment of the ith filter;
the characteristic coefficient vector of each filter is used as a row vector, and all row vectors form an acoustic signal characteristic coefficient matrix;
determining a covariance matrix of the acoustic signal characteristic coefficient matrix, and calculating a principal component information modulus value of the covariance matrix, wherein the method specifically comprises the following steps:
determining the covariance matrix of the characteristic coefficient matrix of the acoustic signal asWherein R is covariance matrix, N is acoustic signal length, F is acoustic signal characteristic coefficient matrix, and T represents matrix transposition;
performing eigenvalue decomposition on the covariance matrix to obtain a plurality of eigenvalues and eigenvectors of the covariance matrix;
based on the eigenvalues and eigenvectors of the covariance matrix, the formula is usedCalculating a principal component information module value of the covariance matrix; wherein ζ is the principal component information module value, λ 1 And lambda (lambda) W Respectively the maximum and minimum eigenvalues in the eigenvalues, |z 1 I represents the feature vector z 1 Is a mold of (a).
2. The method for monitoring the operation state of a transformer based on acoustic signals according to claim 1, characterized in that the determination of the filter bank from the acoustic signals comprises in particular:
performing discrete Fourier transform on the acoustic signal to obtain spectrum distribution of the acoustic signal;
according to the spectrum distribution of the acoustic signal, determining a discrete time response function and a center frequency of each filter in the filter bank as follows:
wherein,as a discrete time response function of the jth filter, n is time, a j A is the scale factor of the jth filter j ＝f L /f cj ，f L Is the lowest center frequency of the filter bank, f H F is the cut-off frequency of the filter bank L And f H Is determined by the frequency spectrum distribution of the transformer acoustic signal; f (f) cj For the center frequency of the jth filter, b is a time shift factor, alpha and beta are respectively a first positive real number and a second positive real number, theta is an initial phase, u () is a unit step function, Q 0 As a quality factor, B 0 For minimum bandwidth, P is the number of filters.
3. An acoustic signal based transformer operating condition monitoring system, the system comprising:
the sound signal acquisition module is used for acquiring sound signals of the transformer in operation by using the microphone sensor;
a filter bank determination module for determining a filter bank from the acoustic signal;
the sound signal characteristic coefficient matrix forming module is used for taking the sound signal as the input of the filter bank, calculating characteristic coefficients in the frequency spectrum range of the sound signal covered by each filter in the filter bank, and forming a sound signal characteristic coefficient matrix;
the principal component information module value calculating module is used for determining a covariance matrix of the acoustic signal characteristic coefficient matrix and calculating a principal component information module value of the covariance matrix;
the normal operation judging module is used for judging that the operation state of the transformer is normal if the principal component information module value is larger than or equal to a preset module value threshold value;
the abnormal operation judging module is used for judging that the operation state of the transformer is abnormal if the principal element information module value is smaller than a preset module value threshold value;
the acoustic signal characteristic coefficient matrix forming module specifically comprises:
characteristic coefficient vector compositionA sub-module for taking the acoustic signal as input of the filter bank and utilizing the formulaCalculating characteristic coefficients in the frequency spectrum range of the acoustic signal covered by each filter in the filter bank to form a characteristic coefficient vector of each filter; wherein F is i (b) Is the characteristic coefficient vector of the ith filter, s (k) is the acoustic signal at the kth moment, N is the acoustic signal length,/for the filter>A discrete time response function corresponding to the acoustic signal at the kth moment of the ith filter;
the sound signal characteristic coefficient matrix forms a sub-module, and the characteristic coefficient vector of each filter is used as a row vector, and all row vectors form the sound signal characteristic coefficient matrix;
the principal component information module value calculating module specifically comprises:
a covariance matrix determination submodule for determining a covariance matrix of the acoustic signal characteristic coefficient matrix asWherein R is covariance matrix, N is acoustic signal length, F is acoustic signal characteristic coefficient matrix, and T represents matrix transposition;
the eigenvalue decomposition sub-module is used for carrying out eigenvalue decomposition on the covariance matrix to obtain a plurality of eigenvalues and eigenvectors of the covariance matrix;
a principal component information modulus value calculation sub-module for utilizing a formula according to a plurality of eigenvalues and eigenvectors of the covariance matrixCalculating a principal component information module value of the covariance matrix; wherein ζ is the principal component information module value, λ 1 And lambda (lambda) W Respectively the maximum and minimum eigenvalues in the eigenvalues, |z 1 I representsFeature vector z 1 Is a mold of (a).
4. The acoustic signal based transformer operating state monitoring system of claim 3, wherein the filter bank determination module specifically comprises:
the frequency spectrum distribution obtaining submodule is used for carrying out discrete Fourier transform on the acoustic signals to obtain frequency spectrum distribution of the acoustic signals;
the discrete time response function determining submodule is used for determining that the discrete time response function and the center frequency of each filter in the filter bank are respectively as follows:
wherein,as a discrete time response function of the jth filter, n is time, a j A is the scale factor of the jth filter j ＝f L /f cj ，f L Is the lowest center frequency of the filter bank, f H F is the cut-off frequency of the filter bank L And f H Is determined by the frequency spectrum distribution of the transformer acoustic signal; f (f) cj For the center frequency of the jth filter, b is a time shift factor, alpha and beta are respectively a first positive real number and a second positive real number, theta is an initial phase, u () is a unit step function, Q 0 As a quality factor, B 0 For minimum bandwidth, P is the number of filters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111434964.3A CN114137444B (en) | 2021-11-29 | 2021-11-29 | Transformer running state monitoring method and system based on acoustic signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111434964.3A CN114137444B (en) | 2021-11-29 | 2021-11-29 | Transformer running state monitoring method and system based on acoustic signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114137444A CN114137444A (en) | 2022-03-04 |
| CN114137444B true CN114137444B (en) | 2024-04-02 |
Family
ID=80389312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111434964.3A Active CN114137444B (en) | 2021-11-29 | 2021-11-29 | Transformer running state monitoring method and system based on acoustic signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114137444B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105699839A (en) * | 2016-01-28 | 2016-06-22 | 云南电网有限责任公司电力科学研究院 | A voltage transformer winding working state detection method and system |
| CN107748314A (en) * | 2017-10-18 | 2018-03-02 | 国网重庆市电力公司北碚供电分公司 | Transformer Faults Analysis system based on sound wave shock detection |
| CN109031014A (en) * | 2017-12-28 | 2018-12-18 | 国网湖北省电力公司宜昌供电公司 | A kind of transformer synthesis reliability assessment and prediction technique based on operation data |
| CN109409762A (en) * | 2018-11-07 | 2019-03-01 | 国网四川省电力公司成都供电公司 | Based on the harmonic contributions division methods for improving data relation analysis |
| CN110286291A (en) * | 2019-07-15 | 2019-09-27 | 广东石油化工学院 | A kind of running state of transformer vibration sound detection method and system using principal component |
| CN110376455A (en) * | 2019-06-26 | 2019-10-25 | 深圳供电局有限公司 | Transformer working state detection method and device, computer equipment and storage medium |
| CN110398647A (en) * | 2019-06-26 | 2019-11-01 | 深圳供电局有限公司 | Transformer state monitoring method |
| CN110458976A (en) * | 2019-07-18 | 2019-11-15 | 浙江大学 | A kind of propeller cavitation condition detection method based on small echo and principal component analysis |
| CN110971295A (en) * | 2018-09-28 | 2020-04-07 | 富士通株式会社 | Optical fiber nonlinear noise monitoring method and device and optical receiver |
| CN111077384A (en) * | 2019-10-17 | 2020-04-28 | 华北电力大学（保定） | Online monitoring method for internal faults of transformer |
| CN111664934A (en) * | 2020-07-05 | 2020-09-15 | 广东石油化工学院 | Transformer state vibration and sound detection signal filtering method and system using feature selection |
| CN112119319A (en) * | 2018-05-18 | 2020-12-22 | 赖茵豪森机械制造公司 | State analysis of an inductively operating component |
| CN112395959A (en) * | 2020-10-30 | 2021-02-23 | 天合云能源互联网技术(杭州)有限公司 | Power transformer fault prediction and diagnosis method and system based on audio features |
-
2021
- 2021-11-29 CN CN202111434964.3A patent/CN114137444B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105699839A (en) * | 2016-01-28 | 2016-06-22 | 云南电网有限责任公司电力科学研究院 | A voltage transformer winding working state detection method and system |
| CN107748314A (en) * | 2017-10-18 | 2018-03-02 | 国网重庆市电力公司北碚供电分公司 | Transformer Faults Analysis system based on sound wave shock detection |
| CN109031014A (en) * | 2017-12-28 | 2018-12-18 | 国网湖北省电力公司宜昌供电公司 | A kind of transformer synthesis reliability assessment and prediction technique based on operation data |
| CN112119319A (en) * | 2018-05-18 | 2020-12-22 | 赖茵豪森机械制造公司 | State analysis of an inductively operating component |
| CN110971295A (en) * | 2018-09-28 | 2020-04-07 | 富士通株式会社 | Optical fiber nonlinear noise monitoring method and device and optical receiver |
| CN109409762A (en) * | 2018-11-07 | 2019-03-01 | 国网四川省电力公司成都供电公司 | Based on the harmonic contributions division methods for improving data relation analysis |
| CN110398647A (en) * | 2019-06-26 | 2019-11-01 | 深圳供电局有限公司 | Transformer state monitoring method |
| CN110376455A (en) * | 2019-06-26 | 2019-10-25 | 深圳供电局有限公司 | Transformer working state detection method and device, computer equipment and storage medium |
| CN110286291A (en) * | 2019-07-15 | 2019-09-27 | 广东石油化工学院 | A kind of running state of transformer vibration sound detection method and system using principal component |
| CN110458976A (en) * | 2019-07-18 | 2019-11-15 | 浙江大学 | A kind of propeller cavitation condition detection method based on small echo and principal component analysis |
| JP2021018818A (en) * | 2019-07-18 | 2021-02-15 | 浙江大学Ｚｈｅｊｉａｎｇ Ｕｎｉｖｅｒｓｉｔｙ | Propeller cavitation state detection method based on wavelet and principal component analysis |
| CN111077384A (en) * | 2019-10-17 | 2020-04-28 | 华北电力大学（保定） | Online monitoring method for internal faults of transformer |
| CN111664934A (en) * | 2020-07-05 | 2020-09-15 | 广东石油化工学院 | Transformer state vibration and sound detection signal filtering method and system using feature selection |
| CN112395959A (en) * | 2020-10-30 | 2021-02-23 | 天合云能源互联网技术(杭州)有限公司 | Power transformer fault prediction and diagnosis method and system based on audio features |
Non-Patent Citations (2)
| Title |
|---|
| 基于变分模式分解和微积分增...能量算子的滚动轴承故障诊断;张东;工程科学学报;第38卷(第9期);1327-1334 * |
| 基于提取振源方向的频谱类型识别;顾煜炯;自动化仪表;第40卷(第2期);1-7 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114137444A (en) | 2022-03-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116295661B (en) | Transformer fault early warning system based on Internet of things | |
| CN106934421B (en) | Power transformer fault detection method based on 2DPCA and SVM | |
| CN110864801A (en) | Transformer noise and vibration on-line monitoring and fault analysis system | |
| CN101930047A (en) | On-line monitoring device of transformer winding state and monitoring method thereof | |
| CN110221137B (en) | Distribution transformer abnormal state detection method based on vibration-sound correlation | |
| CN105699839B (en) | A kind of transformer winding working state detecting method and system | |
| CN108535638B (en) | On-load tap-changer mechanical state monitoring method and system based on multilayer filter | |
| CN112201260A (en) | Transformer running state online detection method based on voiceprint recognition | |
| CN112147444B (en) | Power transformer working state monitoring method and system | |
| CN109443190B (en) | Transient traveling wave-based transformer winding deformation online monitoring method and device | |
| CN107976251A (en) | A kind of transmission pressure structure destroys on-line monitoring system and monitoring method | |
| CN118278624B (en) | Intelligent analysis system for electric power monitoring data based on mutual inductor | |
| CN116184265A (en) | Lightning arrester leakage current detection method and system based on multi-classification SVM | |
| CN115655737A (en) | Health data acquisition and state judgment system for chassis dynamometer equipment | |
| CN114137444B (en) | Transformer running state monitoring method and system based on acoustic signals | |
| Liu et al. | Dry-type AMDT fault detection based on vibration signal analysis by wavelet packet | |
| CN109085442B (en) | Transformer monitoring system based on block chain | |
| CN117116293A (en) | Machine equipment fault diagnosis system in complex sound field environment | |
| CN112327218A (en) | Transformer online monitoring and fault diagnosis method | |
| CN114167316B (en) | Transformer winding mechanical state monitoring method based on manifold distance entropy | |
| CN116884432A (en) | VMD-JS divergence-based power transformer fault voiceprint diagnosis method | |
| CN115639274A (en) | Transformer conservator capsule aging evaluation system and method | |
| CN113900050A (en) | Dry-type transformer state monitoring and fault diagnosis device and system based on vibration signals | |
| CN118197359B (en) | Transformer fault identification method and system based on voiceprint and electromagnetic feature fusion | |
| CN110146772A (en) | A kind of loose defect diagnostic method of transformer core based on rumble spectrum matrix |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |