CN115468751A - Method and device for sound collection and defect identification of transformer - Google Patents
Method and device for sound collection and defect identification of transformer Download PDFInfo
- Publication number
- CN115468751A CN115468751A CN202210958442.1A CN202210958442A CN115468751A CN 115468751 A CN115468751 A CN 115468751A CN 202210958442 A CN202210958442 A CN 202210958442A CN 115468751 A CN115468751 A CN 115468751A
- Authority
- CN
- China
- Prior art keywords
- transformer
- sound
- data
- defect
- box
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention provides a method and a device for sound collection and defect identification of a transformer, wherein the working sound of the transformer in a transformer substation is acquired through a microphone; collecting the obtained transformer working sound and converting the obtained transformer working sound into an output digital signal through converter equipment; uploading the digital signals to a cloud server, and preprocessing the collected digital signals through a processor of the server to obtain sample data; inputting sample data characteristic values of the transformer in different states under different noise conditions into a trained BP neural network, and calculating to obtain a defect code; and obtaining the working state of the transformer according to the defect codes, and giving an alarm in time if abnormal data is obtained. According to the invention, corresponding defects can be obtained for sample data characteristic values of the transformer in different states under different noise conditions, the data transmission device sends data to the cloud server in time, the data is monitored in real time, abnormal data is found, and an alarm is given in time, so that more serious accidents are avoided.
Description
Technical Field
The invention relates to the field of transformer defect detection, in particular to a method and a device for sound collection and defect identification of a transformer.
Background
The power transformer is one of the most important devices in the power system, and the failure of the power transformer can have great influence on the stable and safe operation of the power grid. The transformer has a complex structure, a large volume and large operation and maintenance workload, and the main transformer has a complex power failure process and great difficulty. Therefore, how to improve the identification rate of the potential hazards of the transformer and avoid accidents caused by further development of transformer faults is a current research hotspot. The charged state monitoring and fault diagnosis technology is safe and reliable, and meanwhile, the working efficiency of operation and maintenance personnel can be improved. The on-line method includes a vibration signal analysis method, an on-line reactance method, a pulse signal injection method, and the like. Among them, the vibration signal analysis method has attracted extensive attention because it can sensitively reflect the mechanical state of the winding, is easy to realize the live detection, has no direct electrical connection with the system, and so on. The vibration signal analysis method is one of transformer fault diagnosis methods, and becomes a reliable transformer fault diagnosis technology which is developed rapidly and has good application prospect in recent years. The vibration signal of the transformer contains the fault information of the transformer, but the transformer of the existing detection device is interfered by surrounding electric equipment and noise signals in the actual operation, the accuracy rate of vibration signal identification is not high, and the potential hazard of the transformer cannot be quickly identified. Therefore, it is desirable to provide a method and apparatus for transformer sound collection and defect identification to solve the above problems.
Disclosure of Invention
The invention aims to provide a method and a device for sound collection and defect identification of a transformer, and aims to solve the problems that the transformer of the conventional detection device is interfered by surrounding electric equipment and noise signals in actual operation, the accuracy of vibration signal identification is low, and potential hazards of the transformer cannot be quickly identified.
In one aspect, the present invention provides a method for sound collection and defect identification of a transformer, including:
acquiring the working sound of a transformer in the transformer substation through a microphone;
collecting the obtained transformer working sound and converting the obtained transformer working sound into an output digital signal through converter equipment;
uploading the digital signals to a cloud server, and preprocessing the collected digital signals through a processor of the server to obtain sample data;
inputting sample data characteristic values of the transformer in different states under different noise conditions into a trained BP neural network, and calculating to obtain a defect code;
and obtaining the working state of the transformer according to the defect codes, and giving an alarm in time if abnormal data is obtained.
Further, acquiring transformer operating sound in the substation through a microphone includes:
the MAX9812 microphone module for microphone winnowing;
the equipment used in the acquisition comprises a nine-array element acquisition array consisting of a MAX9812 voice microphone and a notebook computer; the parameters for collecting the sample include: the MIC module array comprises 9 paths, the acquired data is 16bit AD, the sampling rate is 8Hz, and the acquisition time is 3min at the interval of 5 min.
Further, preprocessing the collected digital signals to obtain sample data, including:
the pretreatment comprises the following steps: data filtering, data normalization, data pre-emphasis, framing and windowing;
the data filtering includes: based on the evaluation with rich experience, determining the maximum deflection of the two samples as a group A, and evaluating each time a new value is detected; if the current value differs from the above value by no more than a and the difference between the current value and the next value by no more than a, then the current value is valid, otherwise it is an error;
the data normalization adopts a linear normalization method;
the data pre-emphasis comprises: in the continuous part of the sound signal, the energy of the upper part is usually lower than that of the lower part, and a smaller noise effect in the process of acquiring the sound signal will improve the energy of the lower part, resulting in a higher frequency; a first order filter can achieve significant effects: the example sum dimension is represented as x [ n ], and n is a time index and is a constant, a is in a value range of 0.9 ≦ a ≦ 1.0, and the filter expression in the time domain is: y [ n ] = x [ n ] -ax [ n-1].
Further, inputting the characteristic values of sample data of the transformer in different states under different noise conditions into the trained BP neural network, and calculating to obtain a defect code, wherein the defect code comprises the following steps:
under the noiseless environment, vibration signals under the normal state, the winding loosening defect state and the iron core loosening defect state are respectively measured, and frequency domain analysis is carried out on the measured signals;
according to the vibration signal spectrogram obtained by measurement and calculation, the characteristic quantity can be obtained: fundamental frequency amplitude, frequency multiplication ratio and vibration entropy;
taking the fundamental frequency amplitude, the fundamental frequency specific gravity and the vibration entropy characteristic quantity as neurons of an input layer of the BP neural network, wherein the number of the neurons of the input layer is 3;
determining the number of output layer nerves to be 3 according to the state type of the transformer;
training is carried out on 90 samples without noise interference, wherein the samples without the defects are 30, the samples with loose cores are 30, and the samples with loose windings are 30. The hidden layer is set to be 1, the number of the neurons is 5, the network is iteratively trained through a gradient descent function, the maximum iteration number is set to be 1000, the loss function approaches to 0 when the network is trained for about 300 times, and the network model is converged.
Further, the working state of the transformer is obtained according to the defect codes, and if abnormal data are obtained, an alarm is given in time, and the method comprises the following steps:
and when the abnormal data of the working state of the transformer is obtained according to the coding defects, the abnormal data is uploaded to the cloud end through the server, and the abnormal information is sent to an operator end.
In another aspect, the present invention provides a transformer sound collection and defect identification apparatus, including: the system comprises an outside-box sound collector, an inside-box sound collector, a sound digital signal conversion device and a data transmission device;
the plurality of outside-box sound collectors are arranged outside the transformer box body; the plurality of in-box sound collectors are arranged in the transformer box body and the box body corresponding to the out-box sound collectors; the sound digital signal conversion device is arranged on the left side inside the transformer box body; the data transmission device is arranged in the transformer box body and is positioned above the sound digital signal conversion device; the sound digital signal conversion device is respectively and electrically connected with the sound collector outside the box, the sound collector inside the box and the data transmission device, and the data transmission device is connected with the server through a wireless network so as to acquire the working sound of the transformer in the transformer substation through the microphone; collecting the acquired transformer working sound and converting the acquired transformer working sound into an output digital signal through converter equipment; uploading the digital signals to a cloud server, and preprocessing the collected digital signals through a processor of the server to obtain sample data; inputting sample data characteristic values of the transformer in different states under different noise conditions into a trained BP neural network, and calculating to obtain a defect code; and obtaining the working state of the transformer according to the defect codes, and giving an alarm in time if abnormal data is obtained.
Furthermore, the box outer sound collector is respectively arranged in the middle of the upper surface of the outer part of the transformer box body, the middle of the outer surfaces of the left box body and the right box body and the middle of the outer surface of the box body rear plate.
Furthermore, the in-box sound collector is respectively arranged in the middle of the upper part inside the transformer box body, the middle of the inner surfaces of the left box body and the right box body, the middle of the inner surface of the box body rear plate and the middle of the upper surface of the box body bottom plate.
The invention has the following beneficial effects: the invention provides a method and a device for sound collection and defect identification of a transformer, wherein the method obtains the working sound of the transformer in a transformer substation through a microphone; collecting the acquired transformer working sound and converting the acquired transformer working sound into an output digital signal through converter equipment; uploading the digital signals to a cloud server, and preprocessing the collected digital signals through a processor of the server to obtain sample data; inputting sample data characteristic values of the transformer in different states under different noise conditions into a trained BP neural network, and calculating to obtain a defect code; the working state of the transformer is obtained according to the defect codes, corresponding defects can be obtained according to sample data characteristic values of the transformer in different states under different noise conditions, real-time monitoring of data of the defects of the transformer is achieved, abnormal data are found, an alarm is given in time, and more serious accidents are avoided.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for sound collection and defect identification of a transformer;
FIG. 2 is a front view of the installation distribution of a transformer sound collection and defect identification device;
fig. 3 is a schematic view of the installation distribution of a transformer sound collection and defect identification device.
Illustration of the drawings: 100-an out-of-box sound collector; 200-a sound collector in the box; 300-sound digital signal conversion device; 400-a data transmission device; 500-transformer tank.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the specific embodiments of the present invention and the accompanying drawings. It is to be understood that the disclosed embodiments are merely exemplary of the invention, and are not intended to be exhaustive or exhaustive. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below with reference to the accompanying drawings.
Referring to fig. 1, a method for sound collection and defect identification of a transformer includes:
s101, acquiring transformer working sound in the transformer substation through a microphone.
In this embodiment, the microphone is an MAX9812 microphone module which has strong reliability, high sensitivity, small volume and is easy to install and carry;
the equipment used in the acquisition comprises a nine-array element acquisition array consisting of a MAX9812 voice microphone and a notebook computer; the parameters for collecting the sample include: the MIC module array is 9 paths, the acquired data is 16bit AD, the sampling rate is 8Hz, and the acquisition time is 3min at the interval of 5 min.
S102, collecting the acquired transformer working sound, and converting the acquired transformer working sound into an output digital signal through converter equipment.
S103, preprocessing the collected digital signals to obtain sample data.
In this embodiment, the preprocessing includes: data filtering, data normalization, data pre-emphasis, framing and windowing;
in this embodiment, the data filtering includes: based on the evaluation with rich experience, determining the maximum deflection of the two samples as a group A, and evaluating each time a new value is detected; if the current value differs from the above value by no more than a and the difference between the current value and the next value by no more than a, then the current value is valid, otherwise it is an error;
in this embodiment, the data normalization adopts a linear normalization method.
In this embodiment, the data pre-emphasis includes: in the continuous part of the sound signal, the energy of the upper part is usually lower than that of the lower part, and a smaller noise effect in the process of acquiring the sound signal will improve the energy of the lower part, resulting in a higher frequency; a first order filter can achieve significant effects: if the example sum dimension is represented as x [ n ], and n is a time index and is a constant, a is in a range of 0.9 ≦ a ≦ 1.0, and the filter expression in the time domain is: y [ n ] = x [ n ] -ax [ n-1];
in this embodiment, the framing and windowing includes: the sound signal changes continuously along with the change of time, and is observed on a time axis to be non-stationary, so that the statistical characteristic of the sound signal is not invariable; however, it can be seen that the short-time sound signal is in a relatively steady state; that is, in an extreme time, the sound signal can be regarded as a relatively stable signal, which is stable for a short time; by windowing and framing the sound, a stationary signal can be obtained, the sound signals may not simply intersect on the time axis because of their unique continuity and importance, there is some overlap between the top frame of the current frame and the end of the previous frame, mostly 1/3 or 1/2; this termination will affect the fourier analysis if the signal breaks at the boundary, so a sliding window edge is typically used and the edge is reduced to zero. It has the following characteristics: a cosine window with low side lobes and a wider main lobe:
by using an audio-visual making method based on a balance equation, for example, after audio in any format is dispersed by a special Fourier, the audio-visual making method changes, and the short-term structure STFT of the Fourier can be used for permanent classification verification by using STFT data; the window symbols x n 8230xm of each plate are used as input, the output is the size and the surface of the continuous part of the original symbol, the law of continuous change of sound of each format and the law of amplitude change of auditory sound volume of each structure can be obtained through short-term Fourier transform, and the frequency measurement information of each component part is obtained.
When the length of the window is selected, the resolution of the view will increase as the length of the window increases, but the resolution time will decrease; when the length of the window is short, the resolution of the view will decrease, but the visual resolution solution will also increase the length of the window.
S104, inputting the characteristic values of the sample data of the transformer in different states under different noise conditions into the trained BP neural network, and calculating to obtain the defect code.
In this embodiment, in a noise-free environment, vibration signals in a normal state, a winding loosening defect state, and an iron core loosening defect state are respectively measured, and frequency domain analysis is performed on the measured signals;
according to the vibration signal spectrogram obtained by measurement and calculation, the characteristic quantity can be obtained: fundamental frequency amplitude, frequency multiplication specific weight and vibration entropy;
taking the fundamental frequency amplitude, the fundamental frequency specific gravity and the vibration entropy characteristic quantity as neurons of an input layer of the BP neural network, wherein the number of the neurons of the input layer is 3;
and determining the number of output layer nerves to be 3 according to the state type of the transformer. The transformer has three state types of a normal state (Y1), a winding loose (Y2) and a core loose (Y3), when the transformer is in a yi state, yi =1, yj =0 (i ≠ j), and the output of the neural network is Y = [ Y1, Y2, Y3]. In actual operation, the output of a network training sample is yi =0.9, yj =0.1 (i ≠ j), so that the BP neural network has better generalization capability;
training is carried out on 90 samples without noise interference, wherein the samples without the defects are 30, the samples with loose cores are 30, and the samples with loose windings are 30. The hidden layer is set to be 1, the number of the neurons is 5, the network is iteratively trained through a gradient descent function, the maximum iteration number is set to be 1000, the loss function approaches to 0 when the network is trained for about 300 times, and the network model converges.
S105, obtaining the working state of the transformer according to the defect codes, and giving an alarm in time if abnormal data are obtained.
In this embodiment, when abnormal data of the working state of the transformer is obtained according to the coding defects, the abnormal data is uploaded to the cloud through the server, and the abnormal information is sent to the operator side.
The invention provides a transformer sound collection and defect recognition device, which comprises: the sound collector 100 outside the box, the sound collector 200 inside the box, the sound digital signal conversion device 300 and the data transmission device 400; the outside-box sound collector 100 and the inside-box sound collector 200 may be microphones.
The plurality of outside-box sound collectors 100 are arranged outside the transformer box 500; the plurality of in-box sound collectors 200 are arranged inside the transformer box 500 corresponding to the out-box sound collectors 100; the sound digital signal conversion device 300 is arranged at the left side inside the transformer tank 500; the data transmission device 400 is disposed inside the transformer tank 500 above the sound digital signal conversion device 300. The sound digital signal conversion device 300 is electrically connected to the sound collector 100 outside the box, the sound collector 200 inside the box, and the data transmission device 400, respectively. The data transmission device 400 is connected with a server through a wireless network to realize the method for sound collection and defect identification of the transformer.
Further, the outside-box sound collector 100 is respectively disposed in the middle of the outer upper surface of the transformer box 500, the middle of the outer surfaces of the left and right boxes, and the middle of the outer surface of the box back plate, and is configured to collect sounds outside the box.
Further, the in-box sound collector 200 is respectively disposed in the middle of the upper portion inside the transformer box 500, the middle of the inner surfaces of the left and right boxes, and the middle of the inner surface of the box back plate, for collecting the sound in the box.
The working principle of the device for sound collection and defect identification of the transformer provided by the invention is as follows: the method comprises the steps that the sound of the working state of the transformer is collected completely through an outside-box sound collector 100 distributed outside a transformer box body 500 and an inside-box sound collector 200 distributed inside the transformer box body 500, digital signals of the sound of the working state of the transformer are uploaded to a cloud server, and the collected digital signals are preprocessed through a processor of the server to obtain sample data; inputting sample data characteristic values of the transformer in different states under different noise conditions into the trained BP neural network, and calculating to obtain a defect code; and obtaining the working state of the transformer according to the defect codes, and giving an alarm in time if abnormal data is obtained.
In the BP neural network established through the algorithm, defect codes are obtained through calculation, corresponding defects can be obtained according to sample data characteristic values of the transformer in different states under different noise conditions, the data transmission device 400 sends data to the cloud server in time, the data are monitored in real time, and abnormal data are found to give an alarm in time.
An embodiment of the present invention further provides a storage medium, and a computer program is stored in the storage medium, and when the computer program is executed by a processor, the computer program implements part or all of the steps in each embodiment of the method for sound collection and defect identification of a transformer provided by the present invention. The storage medium may be a magnetic disk, an optical disk, a Read-only memory (ROM) or a Random Access Memory (RAM).
Those skilled in the art will readily appreciate that the techniques of the embodiments of the present invention may be implemented as software plus a required general purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be substantially or partially embodied in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, or the like, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the method according to the embodiments or some parts of the embodiments.
The above-described embodiments of the present invention do not limit the scope of the present invention.
Claims (8)
1. A method for sound collection and defect identification of a transformer is characterized by comprising the following steps:
acquiring the working sound of a transformer in the transformer substation through a microphone;
collecting the acquired transformer working sound and converting the acquired transformer working sound into an output digital signal through converter equipment;
uploading the digital signals to a cloud server, and preprocessing the collected digital signals through a processor of the server to obtain sample data;
inputting sample data characteristic values of the transformer in different states under different noise conditions into a trained BP neural network, and calculating to obtain a defect code;
and obtaining the working state of the transformer according to the defect codes, and timely alarming if abnormal data is obtained.
2. The method for sound collection and defect identification of the transformer according to claim 1, wherein the step of acquiring the working sound of the transformer in the substation through a microphone comprises the steps of:
the MAX9812 microphone module for microphone winnowing;
the equipment used in the acquisition comprises a nine-array element acquisition array consisting of a MAX9812 voice microphone and a notebook computer; the parameters for collecting the sample include: the MIC module array comprises 9 paths, the acquired data is 16bit AD, the sampling rate is 8Hz, and the acquisition time is 3min at the interval of 5 min.
3. The method of claim 1, wherein preprocessing the collected digital signals to obtain sample data comprises:
the pretreatment comprises the following steps: data filtering, data normalization, data pre-emphasis, framing and windowing;
the data filtering includes: based on the evaluation with rich experience, determining the maximum deflection of the two samples as a group A, and evaluating each time a new value is detected; if the current value differs from the above value by no more than a and the difference between the current value and the next value by no more than a, then the current value is valid, otherwise it is an error;
the data normalization adopts a linear normalization method;
the data pre-emphasis comprises: in the continuous part of the sound signal, the energy of the upper part is usually lower than that of the lower part, and a smaller noise effect in the process of acquiring the sound signal will improve the energy of the lower part, resulting in a higher frequency; a first order filter can achieve significant effects: the example sum dimension is represented as x [ n ], and n is a time index and is a constant, a is in a value range of 0.9 ≦ a ≦ 1.0, and the filter expression in the time domain is: y [ n ] = x [ n ] -ax [ n-1].
4. The method of claim 1, wherein the step of inputting the sample data feature values of the transformer in different states under different noise conditions into a trained BP neural network to calculate a defect code comprises:
under the noiseless environment, vibration signals under the normal state, the winding loosening defect state and the iron core loosening defect state are respectively measured, and frequency domain analysis is carried out on the measured signals;
according to the vibration signal spectrogram obtained by measurement and calculation, the characteristic quantity can be obtained: fundamental frequency amplitude, frequency multiplication ratio and vibration entropy;
taking the fundamental frequency amplitude, the fundamental frequency specific gravity and the vibration entropy characteristic quantity as neurons of an input layer of the BP neural network, wherein the number of the neurons of the input layer is 3;
determining the number of output layer nerves to be 3 according to the state type of the transformer;
training 90 samples without noise interference, wherein the samples without noise interference are 30, the samples with loose iron cores are 30, and the samples with loose windings are 30; the hidden layer is set to be 1, the number of the neurons is 5, the network is iteratively trained through a gradient descent function, the maximum iteration number is set to be 1000, the loss function approaches to 0 when the network is trained for about 300 times, and the network model is converged.
5. The method for sound collection and defect identification of a transformer according to claim 1, wherein the working state of the transformer is obtained according to the defect code, and if abnormal data is obtained, an alarm is given in time, and the method comprises the following steps:
and when the abnormal data of the working state of the transformer is obtained according to the coding defects, the abnormal data is uploaded to the cloud end through the server, and the abnormal information is sent to an operator end.
6. The utility model provides a transformer sound collection and defect identification device which characterized in that includes: the system comprises an outside-box sound collector (100), an inside-box sound collector (200), a sound digital signal conversion device (300) and a data transmission device (400);
the plurality of outside-box sound collectors (100) are arranged outside the transformer box body (500); the plurality of in-box sound collectors (200) are arranged in the transformer box body (500) and the box body corresponding to the out-box sound collector (100); the sound digital signal conversion device (300) is arranged on the left side inside the transformer box body (500); the data transmission device (400) is arranged in the transformer box body (500) and is positioned above the sound digital signal conversion device (300);
the sound digital signal conversion device (300) is respectively and electrically connected with the sound collector (100) outside the box, the sound collector (200) inside the box and the data transmission device (400); the data transmission device (400) is connected with the server through a wireless network so as to preprocess the collected digital signals through a processor of the server to obtain sample data; inputting sample data characteristic values of the transformer in different states under different noise conditions into a trained BP neural network, and calculating to obtain a defect code; and obtaining the working state of the transformer according to the defect codes, and timely alarming if abnormal data is obtained.
7. The sound collection and defect recognition device for the transformer according to claim 6, wherein the sound collector (100) outside the transformer box is respectively arranged in the middle of the upper surface of the transformer box (500), the outer surfaces of the left and right box, and the outer surface of the rear plate of the box.
8. The sound collection and defect recognition device for the transformer as claimed in claim 6, wherein the sound collector (200) is respectively disposed at the middle part of the upper part of the inside of the transformer tank (500), the middle parts of the inner surfaces of the left and right tanks, the middle part of the inner surface of the rear plate of the tank, and the middle part of the upper surface of the bottom plate of the tank.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210958442.1A CN115468751A (en) | 2022-08-09 | 2022-08-09 | Method and device for sound collection and defect identification of transformer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210958442.1A CN115468751A (en) | 2022-08-09 | 2022-08-09 | Method and device for sound collection and defect identification of transformer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115468751A true CN115468751A (en) | 2022-12-13 |
Family
ID=84367391
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210958442.1A Pending CN115468751A (en) | 2022-08-09 | 2022-08-09 | Method and device for sound collection and defect identification of transformer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115468751A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117232644A (en) * | 2023-11-13 | 2023-12-15 | 国网吉林省电力有限公司辽源供电公司 | Transformer sound monitoring fault diagnosis method and system based on acoustic principle |
-
2022
- 2022-08-09 CN CN202210958442.1A patent/CN115468751A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117232644A (en) * | 2023-11-13 | 2023-12-15 | 国网吉林省电力有限公司辽源供电公司 | Transformer sound monitoring fault diagnosis method and system based on acoustic principle |
| CN117232644B (en) * | 2023-11-13 | 2024-01-09 | 国网吉林省电力有限公司辽源供电公司 | Transformer sound monitoring fault diagnosis method and system based on acoustic principle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111325095B (en) | Intelligent detection method and system for equipment health state based on acoustic wave signals | |
| CN111442827B (en) | Optical fiber passive online monitoring system for transformer winding vibration | |
| CN112201260B (en) | Transformer running state online detection method based on voiceprint recognition | |
| CN107748314A (en) | Transformer Faults Analysis system based on sound wave shock detection | |
| CN110244204A (en) | A kind of switchgear method for diagnosing faults, system and the medium of multiple characteristic values | |
| CN102707203A (en) | Discriminating and measuring method for partial discharge modes of transformer | |
| CN111912519B (en) | Transformer fault diagnosis method and device based on voiceprint frequency spectrum separation | |
| CN104167207A (en) | Equipment sound identification method based on transformer substation patrol inspection robot | |
| CN104409073A (en) | Substation equipment sound and voice identification method | |
| CN109443190B (en) | Transient traveling wave-based transformer winding deformation online monitoring method and device | |
| US20210364481A1 (en) | Method for evaluating mechanical state of high-voltage shunt reactor based on vibration characteristics | |
| CN110412368A (en) | Electrical equipment online supervision method and system based on Application on Voiceprint Recognition | |
| CN112599134A (en) | Transformer sound event detection method based on voiceprint recognition | |
| CN113551765A (en) | Sound spectrum analysis and diagnosis method for equipment fault | |
| CN115468751A (en) | Method and device for sound collection and defect identification of transformer | |
| CN114217149A (en) | Transformer acoustic fingerprint uninterrupted power detection and state early warning method | |
| CN115687969A (en) | Low-voltage transformer fault diagnosis method based on sound characteristic analysis | |
| CN112052712B (en) | Power equipment state monitoring and fault identification method and system | |
| Zhu et al. | Optimization research on abnormal diagnosis of transformer voiceprint recognition based on improved wasserstein GAN | |
| CN117110744A (en) | Transformer fault diagnosis method and system based on voiceprint analysis | |
| Dang et al. | Cochlear filter cepstral coefficients of acoustic signals for mechanical faults identification of power transformer | |
| CN114157023B (en) | Distribution transformer early warning information acquisition method | |
| CN116298509A (en) | Online identification method for harmonic resonance of power system | |
| CN115410599A (en) | Method and system for identifying running state of transformer based on voiceprint perception | |
| Zou et al. | Research on Power Equipment Fault Diagnosis Technology Based on Acoustic Signal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |