CN111830439B - Transformer fault detection method and transformer - Google Patents
Transformer fault detection method and transformer Download PDFInfo
- Publication number
- CN111830439B CN111830439B CN201910320333.5A CN201910320333A CN111830439B CN 111830439 B CN111830439 B CN 111830439B CN 201910320333 A CN201910320333 A CN 201910320333A CN 111830439 B CN111830439 B CN 111830439B
- Authority
- CN
- China
- Prior art keywords
- transformer
- fault
- vibration signal
- executing
- judging whether
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 33
- 239000013598 vector Substances 0.000 claims abstract description 45
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 24
- 238000001228 spectrum Methods 0.000 claims abstract description 13
- 230000002159 abnormal effect Effects 0.000 claims abstract description 12
- 238000000034 method Methods 0.000 claims description 20
- 238000010438 heat treatment Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 238000013021 overheating Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 3
- -1 amplitude Substances 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000008439 repair process Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000009529 body temperature measurement Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052755 nonmetal Inorganic materials 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
- G01J5/0096—Radiation pyrometry, e.g. infrared or optical thermometry for measuring wires, electrical contacts or electronic systems
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
- G01J2005/0077—Imaging
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Protection Of Transformers (AREA)
Abstract
The invention discloses a transformer fault detection method, which comprises the steps of primarily judging whether a fault occurs according to input current and output current of a transformer; performing wavelet decomposition on the input current of the transformer, and judging whether the input current is abnormal or not according to an obtained decomposition signal; determining whether a sleeve on the high-voltage side of the transformer is abnormal or not according to a display picture monitored by an infrared thermal imager arranged on the periphery of the transformer in a preset continuous time period; judging whether the input current is smaller than the fault judgment starting current or not, and reading the recording information generated in the transformer collected in the set time period when the output current is continuously larger than or equal to the fault judgment starting current at any moment in the set time period; acquiring a characteristic vector of the transformer according to the recording information generated in the transformer; judging whether the vector distance between the time characteristic vector and the known fault characteristic frequency spectrum vector in the characteristic frequency spectrum database is greater than a preset distance threshold value or not; and determining the fault according to the known fault characteristic frequency spectrum.
Description
Technical Field
The invention relates to the field of electric power, in particular to a transformer fault detection method and a transformer.
Background
The high-voltage transformer is a key device of the power system, and the reliable operation of the high-voltage transformer is crucial to the stable operation of the power system. Improving the reliability of the transformer can significantly improve the reliability of the power supply. In order to improve the reliability of the transformer, the current mainstream method is to perform online monitoring on some operation parameters of the transformer according to the national standard, monitor key operation parameters of the transformer, and alarm when the key operation parameters exceed a certain threshold. The method is used for online diagnosis of the fault of the transformer, when the online monitoring system of the transformer gives an alarm, the transformer usually has a remarkable abnormal state or has some faults, and the normal operation and the reliable operation of a power grid are influenced at the moment. In addition, a common method is to perform regular inspection tour and inspection, live detection and state maintenance on each transformer according to national standards, the method is an off-line detection method, compared with on-line monitoring, the method greatly improves the accuracy rate, and sometimes slight abnormality or early failure of the transformer can be found in advance. The live detection and the state detection of the transformer both have longer time intervals, the detection and the repair are carried out once every half year to one year on average, and if the abnormality or the fault occurs between the two time intervals of the detection and the repair of the transformer, the abnormality or the fault cannot be known in advance or avoided by an off-line detection method according to the running condition of the transformer.
Disclosure of Invention
An object of the present invention is to provide a transformer fault detection method and a transformer, which make the fault detection of the transformer more sensitive and efficient.
Specifically, the invention is realized by the following technical scheme:
a method of transformer fault detection, the method comprising the steps of:
s1: whether a fault occurs is primarily judged according to the input current and the output current of the transformer, if yes, S2 is executed, and if not, S1 is continuously executed;
s2: carrying out wavelet packet decomposition on input current of the transformer, judging whether the input current is abnormal or not according to an obtained wavelet packet decomposition signal, if so, executing S3, and if not, executing S9;
s3: determining whether a sleeve on the high-voltage side of the transformer is abnormal or not according to a display picture monitored by a thermal infrared imager arranged on the periphery of the transformer in a preset continuous time period, if so, executing S4, and if not, executing S1;
s4: correcting the intensity of the vibration signal according to the characteristics of the infrared image of the bushing at the high-voltage side of the transformer;
s5: determining corresponding deep structure characteristics according to the multiple characteristic images to obtain corresponding relations between different fault situations of the high-voltage side sleeve of the transformer and the deep structure characteristics;
s6: judging whether the corresponding degree between the corrected vibration signal intensity and the corresponding deep structure characteristic is greater than a preset corresponding degree threshold value or not, if so, executing S7, and otherwise, executing S8;
s7: judging whether a pixel value larger than a preset threshold value exists in a gray level image of an infrared image of a high-voltage side sleeve of the transformer, and if so, acquiring a characteristic image of a transformer component in the transformer to be diagnosed according to the position distribution of the pixel points larger than the preset threshold value in the gray level image;
s8: calling a standard feature vector library Bi, and calculating a difference value rho (A, B) between each Bi value and the wavelet packet decomposition signal feature vector Di; comparing the difference rho (A, B) with an alarm threshold value, judging whether the difference is smaller than the alarm threshold value, and if so, judging that the vibration of the transformer is in a normal range; if the vibration is not less than the preset value, the transformer vibrates abnormally and breaks down, and an alarm indicator lamp is turned on;
s9: judging whether the input current is smaller than the fault judgment starting current of the transformer or not, and reading the recording information generated in the transformer collected in the set time period when the output current is continuously larger than or equal to the fault judgment starting current of the transformer from any moment in the set time period;
s10: acquiring a characteristic vector of the transformer according to the recording information generated in the transformer;
s11: judging whether the vector distance between the time characteristic vector and a known fault characteristic frequency spectrum vector stored in a pre-stored characteristic frequency spectrum database is larger than a preset distance threshold value or not, if so, executing S12, and if not, returning to S2;
s12: and determining the fault according to the known fault characteristic frequency spectrum.
Preferably, the S1 includes:
s1001: acquiring the vibration signal intensity of a transformer vibration signal and the input current and output current value of the transformer;
s1002: and judging whether the input current and the output current of the transformer are equal to the fault judgment starting current within a set time period, if so, executing S2, and if not, executing S1001.
Preferably, the S2 includes:
s21: carrying out j-layer wavelet packet decomposition on the input current signal to obtain wavelet packet decomposition signals Di of 2j frequency bands;
s22: and judging whether the sum of the average power of each decomposition signal Di is larger than a preset average power value or not for each decomposition signal Di, if so, executing S3, and if not, executing S9.
Preferably, the S4 includes:
s41: respectively processing a plurality of infrared images of the high-voltage side sleeve of the transformer to obtain a plurality of characteristic images; the characteristic image comprises a grayscale image;
s42: determining an average pixel threshold of the gray level image according to the gray level image;
s43: and correcting the intensity of the vibration signal according to the average pixel threshold of the gray level image.
Preferably, the S10 includes:
s101: acquiring data of the recording information for N times to form a one-dimensional vector, and normalizing the formed one-dimensional vector to be used as a state vector;
s102: and extracting time characteristic vectors such as amplitude, phase, energy and probability distribution from the state vector to represent the state of the transformer.
Preferably, the S6 includes:
and presetting a corresponding relation table of the deep structure characteristics and the vibration signal intensity, wherein the corresponding degree is the ratio of the corrected vibration signal intensity to the vibration signal intensity in the corresponding relation table of the deep structure characteristics and the vibration signal intensity, when the ratio is between 0.5 and 1.5, determining the ratio as a corresponding value, and if the ratio is not between 0.5 and 1.5, calculating the ratio of the corrected vibration signal intensity to the larger vibration signal intensity recorded in the corresponding relation table.
Preferably, in S1002, the fault determination starting current is set to be greater than 2 times the rated current of the corresponding current.
Preferably, in S42,
the average pixel value of the gray level image is calculated firstly, then the area ratio of the core heating area to the total area of the image is calculated, and the average pixel value is divided by the area ratio to obtain the average pixel threshold value of the gray level image.
Preferably, in S43, an average value of pixels on two diagonal lines of the gray-scale image is first calculated, a ratio of the average pixel threshold of the gray-scale image to the average value of the pixels on the two diagonal lines of the gray-scale image is then calculated, and the ratio is multiplied by the original vibration signal intensity to obtain the corrected vibration signal intensity.
A transformer uses the aforementioned transformer fault detection method.
The beneficial effects of the invention are: the fault detection of the transformer is more sensitive and efficient.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic diagram of a transformer fault detection method according to the present invention;
FIG. 2 is a schematic view of a detailed process of S1 in FIG. 1;
FIG. 3 is a schematic view of a detailed process of S2 in FIG. 1;
FIG. 4 is a schematic view of a detailed process of S4 in FIG. 1;
fig. 5 is a schematic diagram of a specific flow of S10 in fig. 1.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the invention, as detailed in the appended claims.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present invention. The word "if," as used herein, may be interpreted as "at … …" or "at … …" or "in response to a determination," depending on the context.
The present invention will be described in detail below by way of examples.
The invention provides a transformer fault detection method, which comprises the following steps:
s1: and primarily judging whether a fault occurs according to the input current and the output current of the transformer, if so, executing S2, and if not, continuing to execute S1.
S2: and carrying out wavelet packet decomposition on the input current of the transformer, judging whether the input current is abnormal according to the obtained wavelet packet decomposition signal, if so, executing S3, and if not, executing S9.
S3: and in a preset continuous time period, determining whether the high-voltage side sleeve of the transformer is abnormal or not according to a display picture monitored by an infrared thermal imager arranged on the periphery of the transformer, if so, executing S4, and otherwise, executing S1.
The infrared temperature measurement technology is applied to the live-line detection of power equipment, is one of routine live-line detection items of the transformer, and can successfully detect various heating defects of the transformer, including dielectric loss heating of a high-voltage bushing, abnormal oil level in the bushing, poor contact of a bushing joint, body leakage heating and the like. The infrared temperature measurement is realized by an infrared thermal imager.
Specifically, the temperature of three high-voltage side bushings of a transformer is determined, and the high-voltage side bushing with the highest temperature is determined according to the temperature of the three high-voltage side bushings of the transformer; and if the difference value between the temperature of the high-pressure side sleeve with the highest temperature and the temperature of any other high-pressure side sleeve is greater than the temperature of the other high-pressure side sleeve, determining that the high-pressure side sleeve with the highest temperature is abnormal.
S4: and correcting the intensity of the vibration signal according to the characteristics of the infrared image of the bushing on the high-voltage side of the transformer.
S5: and determining corresponding deep structure characteristics according to the multiple characteristic images to obtain the corresponding relation between different fault situations of the high-voltage side sleeve of the transformer and the deep structure characteristics.
After acquiring a plurality of characteristic images corresponding to each transformer high-voltage side bushing fault, in this step, deep structure characteristics of the plurality of characteristic images corresponding to each transformer high-voltage side bushing fault are determined, and the deep structure characteristics can reflect appearance characteristics, texture characteristics, fault positions and other characteristics in the plurality of characteristic images.
Specifically, a plurality of characteristic images corresponding to each transformer high-voltage side bushing fault are obtained, and the corresponding deep structure characteristics can be obtained by adopting a convolutional neural network method, for example, a google lenet deep learning framework can be specifically adopted, and the method adopting the convolutional neural network is only a specific implementation manner of the present application, and the present embodiment does not limit the specific implementation manner.
S6: and judging whether the correspondence between the corrected vibration signal intensity and the corresponding deep structure features is greater than a preset correspondence threshold, if so, executing S7, and otherwise, executing S8.
The correspondence, that is, the matching degree between the corrected vibration signal intensity and the deep-layer structural feature, is a pre-quantified numerical value, that is, the correspondence between the deep-layer structural feature and the vibration signal intensity, which is obtained through experiments under the high-voltage side bushing faults of various transformers when the transformers leave a factory. When the correspondence table between the deep structure characteristics and the vibration signal intensity is prepared, the actually detected vibration signal intensity is collected, and since the larger the vibration signal intensity is, the more likely the transformer fails, the corrected vibration signal intensity is adopted when the determination of step S10 is performed, thereby increasing the sensitivity of the determination. Specifically, in the table of correspondence between the deep structure characteristic and the vibration signal intensity, the correspondence between the specific vibration signal intensity and the deep structure characteristic is recorded, the correspondence may be a ratio of the corrected vibration signal intensity to the vibration signal intensity in the table of correspondence between the deep structure characteristic and the vibration signal intensity, when the ratio is between 0.5 and 1.5, the ratio is determined as a correspondence value, and if the ratio is not between 0.5 and 1.5, the ratio is calculated between the corrected vibration signal intensity and the larger vibration signal intensity recorded in the table of correspondence.
S7: and judging whether a pixel value larger than a preset threshold value exists in a gray level image of the transformer high-voltage side sleeve infrared image, if so, acquiring a characteristic image of a transformer component in the transformer to be diagnosed according to the position distribution of the pixel points larger than the preset threshold value in the gray level image.
Since each pixel value in the infrared image of the transformer reflects the temperature of the corresponding position in the transformer, meanwhile, there is a corresponding relationship between the pixel value of the infrared image and the pixel value of the gray scale image corresponding to the infrared image, that is, the pixel value in the infrared image representing a certain temperature range corresponds to the pixel value in the gray scale image in a certain gray scale range. Therefore, in this embodiment, a threshold corresponding to the lowest temperature of the infrared image of each faulty transformer component in the grayscale image is determined, and this threshold is referred to as a preset threshold, and according to the preset threshold, it can be determined whether a pixel value greater than the preset threshold exists in the grayscale image, and if not, it indicates that there is no position where the temperature is higher than the fault temperature in the transformer to be diagnosed, that is, there is no overheating fault in the transformer to be diagnosed; on the contrary, it indicates that there is a location in the transformer to be diagnosed where the temperature is higher than the fault temperature, i.e., there is an overheating fault in the transformer to be diagnosed.
S8: calling a standard feature vector library Bi, and calculating a difference value rho (A, B) between each Bi value and the wavelet packet decomposition signal feature vector Di; comparing the difference rho (A, B) with an alarm threshold value, judging whether the difference is smaller than the alarm threshold value, and if so, judging that the vibration of the transformer is in a normal range; if not, the transformer vibrates abnormally, a fault occurs, and an alarm indicator lamp is turned on.
S9: and judging whether the input current is smaller than the fault judgment starting current of the transformer or not, and reading the recording information generated in the transformer collected in the set time period when the output current is continuously larger than or equal to the fault judgment starting current of the transformer from any moment in the set time period.
S10: and acquiring the characteristic vector of the transformer according to the recording information generated in the transformer.
S11: and judging whether the vector distance between the time characteristic vector and the known fault characteristic frequency spectrum vector stored in the pre-stored characteristic frequency spectrum database is greater than a preset distance threshold, if so, executing S12, and if not, returning to S2.
If the vector distance is greater than the preset distance threshold, it is indicated that the difference between the transformer state corresponding to the time characteristic vector and the transformer state corresponding to the normal working characteristic frequency spectrum vector is large, it is indicated that the current state belongs to a predefined fault state, and a new fault state needs to be determined. If the vector distance is smaller than the preset distance threshold value, the current transformer fault belongs to a predefined fault state.
S12: and determining the fault according to the known fault characteristic frequency spectrum.
By adopting the transformer fault detection method provided by the embodiment of the invention, the source of the transformer fault can be obtained from multiple dimensions in a comprehensive judgment mode, and the fault characteristics are compared, so that the fault state of the transformer can be more accurately obtained.
Further, the S1 includes:
s1001: and acquiring the vibration signal intensity of the vibration signal of the transformer and the input current and the output current value of the transformer.
According to the transformer fault detection method, the intensity of the vibration signal is corrected according to the output current value of the transformer and the preset rated output current value of the transformer, and whether the transformer has a fault or hidden trouble is judged according to the intensity of the corrected vibration signal. In the operation process of the transformer, the voltage basically keeps unchanged, but the output current of the transformer, namely the working current of the transformer, can change along with the change of the load driven by the transformer, and the magnitude of the output current has an influence on the vibration signal, so that the influence brought by the output current of the transformer is brought in the intensity of the vibration signal obtained in the way, and the change of the mechanical structure of the transformer winding cannot be accurately represented. Based on the above, the present embodiment considers the influence of different loads on the output current of the transformer, and corrects the vibration signal intensity according to the transformer output current value and the rated current value, so that the corrected vibration signal intensity can more accurately represent the state of the transformer, and thus the diagnosis result is more accurate. The method provided by the embodiment can accurately diagnose the state of the transformer while the transformer is put into operation, so that workers can find and repair the fault or the hidden fault trouble of the transformer in time, and the stable operation of the power system is ensured.
S1002: and judging whether the input current and the output current of the transformer are equal to the fault judgment starting current within a set time period, if so, executing S2, and if not, executing S1001.
When a line in the transformer fails, the input current of the transformer is generally normal, and the output current of the transformer is often abnormal. In the scheme, the lines at the positions where the current needs to be collected are all provided with the fault judgment starting current corresponding to the line current, and the corresponding fault judgment starting currents on the lines at each position are different. The failure determination starting current mentioned in the present scheme is set to be greater than 2 times its rated current corresponding to the current. When the transformer works, if more electric equipment is connected to the transformer, the phenomenon of overload can occur, and at the moment, the power and the working current of the transformer can fluctuate within a small range under a rated condition. If the input current and the output current are equal to the fault judgment starting current within the set time period, carrying out next fault judgment, if not, judging that the current transformer has no fault condition, and returning to S1.
Further, the S2 includes:
s21: and carrying out j-layer wavelet packet decomposition on the input current signal to obtain wavelet packet decomposition signals Di of 2j frequency bands.
S22: and judging whether the sum of the average power of each decomposition signal Di is larger than a preset average power value or not for each decomposition signal Di, if so, executing S3, and if not, executing S9.
Each of the resolved signals Di represents the current condition of the input current signal at each level, the average power of each resolved signal Di of the input current signal represents the signal strength of the current at each level, and the average power of the resolved signals Di can be extracted as the signal characteristic of the input current signal.
Further, the S4 includes:
s41: respectively processing a plurality of infrared images of the high-voltage side sleeve of the transformer to obtain a plurality of characteristic images; the characteristic image comprises a grayscale image.
The infrared image of the transformer can reflect the temperature of the transformer, and has the characteristics of being free from illumination, electromagnetic signal interference and the like, so that the infrared image of the transformer is widely applied to a method for detecting whether the transformer has an overheating fault or not.
Generally, a plurality of infrared images of a transformer with an overheat fault and a plurality of infrared images of a transformer which normally works are respectively obtained; acquiring a plurality of gray level images of the transformer with the overheat fault to obtain a plurality of first gray level images, and acquiring a plurality of gray level images of the transformer which normally works to obtain a plurality of second gray level images; respectively determining pixel distribution characteristics of the multiple first gray level images and the multiple second gray level images to obtain two pixel distribution characteristics; and determining the corresponding relation between the two distribution characteristics and the overheating fault transformer and the normally working transformer, and determining whether the transformer to be diagnosed has the overheating fault or not based on the corresponding relation and the pixel distribution characteristics of the gray level image of the transformer to be diagnosed.
After obtaining a plurality of infrared images of each fault transformer component, carrying out gray level processing on the corresponding plurality of infrared images to obtain a plurality of gray level images aiming at each fault transformer component, and taking the plurality of gray level images as a plurality of characteristic images.
In practical applications, the denoising processing may be performed on the multiple gray scale images corresponding to each transformer component, specifically, a wavelet transform manner may be adopted, and of course, the denoising processing performed on the multiple gray scale images corresponding to each transformer component is not an action necessarily performed in this embodiment.
S42: an average pixel threshold for the grayscale image is determined from the grayscale image.
Since each pixel value in the infrared image of the transformer reflects the temperature of the corresponding position in the transformer, meanwhile, there is a corresponding relationship between the pixel value of the infrared image and the pixel value of the gray scale image corresponding to the infrared image, that is, the pixel value in the infrared image representing a certain temperature range corresponds to the pixel value in the gray scale image in a certain gray scale range. Therefore, in this embodiment, a threshold corresponding to the lowest temperature of the infrared image of each faulty transformer component in the grayscale image is determined, and this threshold is referred to as a preset threshold, and according to the preset threshold, it can be determined whether a pixel value greater than the preset threshold exists in the first grayscale image, and if not, it indicates that there is no position where the temperature is higher than the fault temperature in the transformer to be diagnosed, that is, there is no overheating fault in the transformer to be diagnosed; on the contrary, it indicates that there is a location in the transformer to be diagnosed where the temperature is higher than the fault temperature, i.e., there is an overheating fault in the transformer to be diagnosed.
Specifically, an average pixel value of the gray level image is calculated first, then, an area ratio of the core heating area to the total area of the image is calculated, and the average pixel value is divided by the area ratio to obtain an average pixel threshold of the gray level image. By adopting the method, the average pixel of the gray level image is inversely proportionally expanded according to the area ratio of the core heating area to the total area of the image, so that the obtained average pixel threshold of the gray level image can represent the size of the starting heat.
S43: and correcting the intensity of the vibration signal according to the average pixel threshold of the gray level image.
Specifically, the average value of the pixels on the two diagonal lines of the gray scale image is calculated, then the ratio of the average pixel threshold value of the gray scale image to the average value of the pixels on the two diagonal lines of the gray scale image is calculated, and the ratio is multiplied by the original vibration signal intensity to obtain the corrected vibration signal intensity. Since the gray scale image of the transformer high-voltage side bushing is generally uniformly changed, the average value of the pixels on the two diagonal lines of the gray scale image of the normally working transformer high-voltage side bushing is approximately equal to the average pixel threshold of the gray scale image, so after the ratio is multiplied by the original vibration signal intensity, if the transformer works normally, the corrected vibration signal intensity is not changed too much, and if the transformer fails, the corrected vibration signal intensity is changed greatly.
Further, the S10 includes:
s101: and carrying out N times of data acquisition on the recording information to form a one-dimensional vector, and normalizing the formed one-dimensional vector to be used as a state vector.
The number of the data points in the N and the decomposition signal Di is consistent.
S102: and extracting time characteristic vectors such as amplitude, phase, energy and probability distribution from the state vector to represent the state of the transformer.
And taking each numerical value in the state vector as a discrete digital signal numerical value, and acquiring time characteristic vectors such as amplitude, phase, energy, probability distribution and the like corresponding to the state vector by using the conventional digital signal processing method.
The transformer fault is the result of the comprehensive action and long-term accumulation of the transformer and the application environment thereof, so that the symptoms of the transformer fault are various, and the connection between the symptoms of the fault and the fault mechanism is also complicated, thereby causing great difficulty in establishing a universal transformer fault control method. According to the invention, after the metal and nonmetal content data and the failure times of the inner sleeve are collected for a long time and rearranged, a state vector uniquely corresponding to the working state of the transformer is formed, and the current failure state of the transformer can be determined from the aspect that the transformer failure is most easily generated.
The invention also provides a transformer using the transformer fault detection method.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. A method of transformer fault detection, the method comprising the steps of:
s1: primarily judging whether a fault occurs according to the input current and the output current of the transformer, if so, executing S2, and if not, continuing to execute S1;
s2: carrying out wavelet packet decomposition on input current of the transformer, judging whether the input current is abnormal or not according to an obtained wavelet packet decomposition signal, if so, executing S3, and if not, executing S9;
s3: determining whether a sleeve on the high-voltage side of the transformer is abnormal or not according to a display picture monitored by a thermal infrared imager arranged on the periphery of the transformer in a preset continuous time period, if so, executing S4, and if not, executing S1;
s4: correcting the intensity of the vibration signal according to the characteristics of the infrared image of the bushing on the high-voltage side of the transformer;
s5: determining corresponding deep structure characteristics according to the multiple characteristic images to obtain corresponding relations between different fault situations of the high-voltage side sleeve of the transformer and the deep structure characteristics;
s6: judging whether the corresponding degree between the corrected vibration signal intensity and the corresponding deep structure characteristic is greater than a preset corresponding degree threshold value or not, if so, executing S7, and otherwise, executing S8;
s7: judging whether a pixel value larger than a preset threshold value exists in a gray level image of an infrared image of a high-voltage side sleeve of the transformer, and if so, acquiring a characteristic image of a transformer component in the transformer to be diagnosed according to the position distribution of the pixel points larger than the preset threshold value in the gray level image;
s8: calling a standard feature vector library Bi, and calculating a difference value rho (A, B) between each Bi value and the wavelet packet decomposition signal feature vector Di; comparing the difference rho (A, B) with an alarm threshold value, judging whether the difference is smaller than the alarm threshold value, and if so, judging that the vibration of the transformer is in a normal range; if the vibration is not less than the preset value, the transformer vibrates abnormally and breaks down, and an alarm indicator lamp is turned on;
s9: judging whether the input current is smaller than the fault judgment starting current of the transformer or not, and reading the recording information generated in the transformer collected in the set time period when the output current is continuously larger than or equal to the fault judgment starting current of the transformer from any moment in the set time period;
s10: acquiring a time characteristic vector of the transformer according to recording information generated in the transformer;
s11: judging whether the vector distance between the time characteristic vector and a known fault characteristic frequency spectrum vector stored in a pre-stored characteristic frequency spectrum database is larger than a preset distance threshold value or not, if so, executing S12, and if not, returning to S2;
s12: and determining the fault according to the known fault characteristic frequency spectrum.
2. The transformer fault detection method of claim 1, wherein the S1 comprises:
s1001: acquiring the vibration signal intensity of a transformer vibration signal and the input current and output current value of the transformer;
s1002: and judging whether the input current and the output current of the transformer are equal to the fault judgment starting current within a set time period, if so, executing S2, and if not, executing S1001.
3. The transformer fault detection method of claim 1, wherein the S2 comprises:
s21: carrying out j-layer wavelet packet decomposition on the input current signal to obtain wavelet packet decomposition signals Di of 2j frequency bands;
s22: and judging whether the sum of the average power of each decomposition signal Di is larger than a preset average power value or not for each decomposition signal Di, if so, executing S3, and if not, executing S9.
4. The transformer fault detection method of claim 1, wherein the S4 comprises:
s41: respectively processing a plurality of infrared images of the high-voltage side sleeve of the transformer to obtain a plurality of characteristic images; the characteristic image comprises a grayscale image;
s42: determining an average pixel threshold of the gray level image according to the gray level image;
s43: and correcting the intensity of the vibration signal according to the average pixel threshold of the gray level image.
5. The transformer fault detection method of claim 1, wherein the S10 comprises:
s101: acquiring data of N times of the recording information to form a one-dimensional vector, and normalizing the formed one-dimensional vector to be used as a state vector;
s102: and extracting amplitude, phase, energy and probability distribution time characteristic vectors from the state vectors for representing the state of the transformer.
6. The transformer fault detection method of claim 1, wherein the S6 comprises:
presetting a corresponding relation table of the deep structure characteristics and the vibration signal strength, wherein the corresponding degree is the ratio of the corrected vibration signal strength to the vibration signal strength in the corresponding relation table of the deep structure characteristics and the vibration signal strength, determining the ratio as a corresponding value when the ratio is between 0.5 and 1.5, and calculating the ratio of the corrected vibration signal strength to the larger vibration signal strength recorded in the corresponding relation table if the ratio is not between 0.5 and 1.5.
7. The transformer fault detection method according to claim 2, wherein in S1002, the fault determination starting current is set to be greater than 2 times the rated current of the corresponding current.
8. The transformer fault detection method of claim 4, wherein, in S42,
the average pixel value of the gray level image is calculated firstly, then the area ratio of the core heating area to the total area of the image is calculated, and the average pixel value is divided by the area ratio to obtain the average pixel threshold value of the gray level image.
9. The transformer fault detection method according to claim 4, wherein in S43, an average value of pixels on two diagonal lines of the gray scale image is calculated, a ratio of a threshold of an average pixel of the gray scale image to an average value of pixels on two diagonal lines of the gray scale image is calculated, and the ratio is multiplied by the original vibration signal intensity to obtain the corrected vibration signal intensity.
10. A transformer using the transformer fault detection method of any one of claims 1~9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910320333.5A CN111830439B (en) | 2019-04-19 | 2019-04-19 | Transformer fault detection method and transformer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910320333.5A CN111830439B (en) | 2019-04-19 | 2019-04-19 | Transformer fault detection method and transformer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111830439A CN111830439A (en) | 2020-10-27 |
| CN111830439B true CN111830439B (en) | 2022-10-11 |
Family
ID=72911699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910320333.5A Active CN111830439B (en) | 2019-04-19 | 2019-04-19 | Transformer fault detection method and transformer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111830439B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112132911B (en) * | 2020-11-25 | 2021-03-16 | 恒银金融科技股份有限公司 | Threshold self-adaption method for infrared digital image detection |
| CN113176522A (en) * | 2021-04-16 | 2021-07-27 | 国网江苏省电力有限公司南通供电分公司 | Transformer short-circuit fault detection method |
| CN113723268B (en) * | 2021-08-25 | 2024-01-26 | 国网北京市电力公司 | Method, device, computer storage medium and processor for identifying power grid faults |
| CN115792729A (en) * | 2022-11-30 | 2023-03-14 | 广东粤电科试验检测技术有限公司 | Transformer composite fault diagnosis method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201210969D0 (en) * | 2011-06-27 | 2012-08-01 | Gen Electric | Electrical substation fault monitoring and diagnostics |
| CN103513139A (en) * | 2013-10-16 | 2014-01-15 | 王丽华 | Power transformer fault intelligent diagnosis technology, method and device |
| CN104898013A (en) * | 2015-06-09 | 2015-09-09 | 北京联合大学 | Method and system for diagnosing circuit fault based on acoustical measurement |
| CN106707060A (en) * | 2016-12-16 | 2017-05-24 | 中国电力科学研究院 | Method for acquiring discrete state parameters of power transformer |
-
2019
- 2019-04-19 CN CN201910320333.5A patent/CN111830439B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201210969D0 (en) * | 2011-06-27 | 2012-08-01 | Gen Electric | Electrical substation fault monitoring and diagnostics |
| CN103513139A (en) * | 2013-10-16 | 2014-01-15 | 王丽华 | Power transformer fault intelligent diagnosis technology, method and device |
| CN104898013A (en) * | 2015-06-09 | 2015-09-09 | 北京联合大学 | Method and system for diagnosing circuit fault based on acoustical measurement |
| CN106707060A (en) * | 2016-12-16 | 2017-05-24 | 中国电力科学研究院 | Method for acquiring discrete state parameters of power transformer |
Non-Patent Citations (1)
| Title |
|---|
| 大型电力变压器过热性故障诊断与处理;林礼清;《电网技术》;19991130;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111830439A (en) | 2020-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111830439B (en) | Transformer fault detection method and transformer | |
| KR102342476B1 (en) | System and method for determining situation of facility by imaging seinsing data of facility | |
| CN110910383B (en) | Phase modulator fault monitoring method and system based on infrared image recognition | |
| CN106950472B (en) | insulator detection method based on infrared and ultraviolet imaging | |
| US20130096853A1 (en) | Systems and methods for monitoring electrical contacts | |
| CN106597160B (en) | Electronic equipment fault detection method and device | |
| CN111830438B (en) | Transformer fault detection method and transformer | |
| CN107563263A (en) | Video image identification method and system based on instrument | |
| CN112906576B (en) | State identification method, device and equipment of isolating switch equipment and temperature measurement method thereof | |
| CN117129790A (en) | Fault diagnosis system for power system | |
| WO2019100030A1 (en) | Remote monitoring of electrical equipment with multi-factor historical / statistical anomaly detection and prediction | |
| JP2017142622A (en) | Monitoring device and monitoring device control method | |
| CN116308300B (en) | Power equipment state monitoring evaluation and command method and system | |
| CN117706943B (en) | Self-adaptive control method and system for converter transformer valve side sleeve end screen voltage divider | |
| CN116226766A (en) | High-voltage electrical apparatus running state monitoring system | |
| CN111221996B (en) | Instrument screen vision detection method and system | |
| CN109725220B (en) | Detection method, system and device for transformer oil cooling loop | |
| CN117590278A (en) | Transformer fault detection method, device, equipment and medium based on power cabinet | |
| CN115908407B (en) | Power equipment defect detection method and device based on infrared image temperature value | |
| CN116304798A (en) | Partial discharge type identification method, device, equipment and medium | |
| CN115754507A (en) | Transformer fault diagnosis method, device and equipment based on vibration signals | |
| CN116188505A (en) | Insulation power equipment on-line monitoring system based on ultraviolet and infrared collaborative detection | |
| CN116381419B (en) | Transmission line fault processing method, device, computer equipment and storage medium | |
| TW201118317A (en) | Vision-based method for combustion process monitoring, diagnosis, and computer-readable medium using the same | |
| CN110399911A (en) | A kind of transformer mechanical failure diagnostic method based on sound-filed simulation cloud atlas |
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 | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 315031 No.17 and 22 Fenglin Road, Cicheng Town, Jiangbei District, Ningbo City, Zhejiang Province Patentee after: Ningbo Aokes Intelligent Technology Co.,Ltd. Address before: 315031 No.17 and 22 Fenglin Road, Cicheng Town, Jiangbei District, Ningbo City, Zhejiang Province Patentee before: NINGBO AUX HIGH TECHNOLOGY Co.,Ltd. |