CN114171058A - Transformer running state monitoring method and system based on voiceprint - Google Patents
Transformer running state monitoring method and system based on voiceprint Download PDFInfo
- Publication number
- CN114171058A CN114171058A CN202111467440.4A CN202111467440A CN114171058A CN 114171058 A CN114171058 A CN 114171058A CN 202111467440 A CN202111467440 A CN 202111467440A CN 114171058 A CN114171058 A CN 114171058A
- Authority
- CN
- China
- Prior art keywords
- audio data
- transformer
- abnormal
- data
- original audio
- 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
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012544 monitoring process Methods 0.000 title claims abstract description 22
- 230000008569 process Effects 0.000 claims abstract description 17
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- 230000002159 abnormal effect Effects 0.000 claims description 87
- 238000004422 calculation algorithm Methods 0.000 claims description 18
- 238000001228 spectrum Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 7
- 238000012512 characterization method Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 4
- 238000009432 framing Methods 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 claims description 2
- 238000004590 computer program Methods 0.000 claims description 2
- 238000012937 correction Methods 0.000 claims description 2
- 238000012986 modification Methods 0.000 claims description 2
- 230000004048 modification Effects 0.000 claims description 2
- 238000005457 optimization Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
- G10L25/18—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being spectral information of each sub-band
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
- G10L25/24—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being the cepstrum
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Protection Of Transformers (AREA)
Abstract
The invention discloses a transformer running state monitoring method and system based on voiceprint, which comprises the following steps: s100, acquiring original audio data of a transformer; s200, determining whether the tone quality of the acquired original audio data needs to be corrected or not according to the tone quality of the original audio data acquired by the transformer; s300, if necessary, correcting based on the most value of the original audio data to obtain effective original audio data; through setting up first, two default, carry out tone quality division to the original audio data that the transformer gathered to ensure that the transformer can gather effectual original audio data, so that carrying out subsequent further analysis in-process, can filter by a wide margin the interference item, reduce analysis process's calculated amount.
Description
Technical Field
The invention relates to the technical field of power equipment, in particular to a transformer running state monitoring method and system based on voiceprints.
Background
Nowadays, the power demand is continuously promoted, the quality problem of electric energy is concerned, and the stable supply relationship of electric energy is national economy. A large number of practices show that latent faults often exist before serious accidents happen to a power transformer (reactor), and main faults of the power transformer (reactor) are caused by accumulation of the latent faults along with time, such as internal partial discharge, local overheating, winding deformation, loosening of mechanical parts, ageing of equipment insulation and the like. Once a fault occurs, inconvenience or economic loss is brought to life. Nowadays, a power transformer (reactor) protection method mainly performs relay protection by electric parameters such as voltage and current during fault, and related latent faults are difficult to detect due to ubiquitous faults.
In the prior art, the operation condition of the transformer can be judged by collecting the sound of the transformer and carrying out voiceprint analysis on the sound. However, most of the sound collected by the transformer is directly subjected to feature extraction and further processing, the calculation amount is large, and a huge data set causes more errors. Meanwhile, the used sample data is not well utilized further, so that the resource waste of computing power is caused, and the sustainable operation of equipment is not facilitated.
Disclosure of Invention
In order to solve the defects in the prior art, the invention aims to overcome the existing defects and provides a transformer running state monitoring method based on voiceprint, which comprises the following steps:
acquiring original audio data of a transformer;
determining whether the tone quality of the acquired original audio data needs to be corrected or not according to the tone quality of the original audio data acquired by the transformer;
if necessary, a correction is made based on the most significant value of the original audio data to obtain valid original audio data.
As a further optimization of the above scheme, the modifying based on the most significant value of the original audio data includes:
judging whether the tone quality of the original audio data is greater than or equal to a first preset value or not;
if so, determining that the tone quality of the collected original audio data needs to be corrected;
if not, determining that the tone quality of the collected original audio data does not need to be corrected.
As a further optimization of the above scheme, judging whether the tone quality of the original audio data is less than or equal to a second preset value;
if so, determining that the tone quality of the collected original audio data needs to be corrected;
if not, determining that the tone quality of the collected original audio data does not need to be corrected.
As a further optimization of the above scheme, the method further classifies the failure of the original audio data, specifically including the following steps:
acquiring target audio data of the transformer based on the original audio data;
if the audio data of the checking target does not exceed the set audio abnormal threshold, marking the current audio as a normal audio;
if the audio data of the checking target exceeds a set audio abnormal threshold, marking the current audio as an abnormal audio, and storing the abnormal audio in an abnormal sample library;
comparing the abnormal audio confirmed by the calibration with an abnormal sample library based on a Zero-shot Learning algorithm:
if the detected abnormal audio type exists in the abnormal sample library, classifying the detected abnormal audio type into the category of the abnormal sample library;
and if the detected abnormal audio type does not exist in the abnormal sample library, classifying the detected abnormal audio type into an unknown abnormal type.
As a further optimization of the above scheme, the unknown abnormal category audio data is labeled and classified based on an Active Learning algorithm, and the updated abnormal sample library is:
recording the original abnormal sample library as SQL 1, recording the times of classifying any target audio Data into the abnormal sample library or unknown abnormal class as n, and recording the target audio Data corresponding to the abnormal sample library as Data a1，Data a2，…Data axMemory for recordingRecording target audio Data corresponding to unknown abnormity as Data b1，Data b2，…Data byWherein n is x + y;
the nth exception sample library is:
SQL n＝SQL 1+Data a1+Data a2+…Data ax+Data b1+…+Data by (1)
as a further optimization of the above scheme, the process of converting the target audio data based on the original audio collected by the transformer specifically includes the following steps:
acquiring original audio data of a transformer environment;
pre-emphasis processing, framing and windowing are carried out on the collected original audio data, and a plurality of pre-processed audios with frames as units are generated;
acquiring a plurality of preprocessed audio frequencies, and acquiring frequency spectrum information corresponding to the preprocessed audio frequencies based on fast Fourier transform;
acquiring Mel frequency spectrum from a plurality of frequency spectrum information by adopting a Mel filter bank;
cepstrum analysis is performed on top of the Mel spectrum to obtain Mel-frequency cepstral coefficients MFCC.
As a further optimization of the above scheme, the cepstrum analysis process specifically includes the following steps:
collecting a plurality of Mel frequency spectrums for logarithmic operation;
implementing an inverse transform by a discrete cosine transform;
and selecting the coefficient after discrete cosine transform as an MFCC coefficient to generate target audio data.
As a further optimization of the above solution, after obtaining a plurality of MFCC coefficients, the generating of the audio anomaly threshold value includes the following steps:
compressing a plurality of MFCC coefficients based on an Auto-encoder algorithm to generate and output a space characterization number;
converting a plurality of said MFCC coefficients into statistical distribution parameters-mean and standard deviation based on a VAE algorithm;
constructing a power equipment voiceprint model library based on the spatial representation number output by the Auto-encoder algorithm and the mean value and standard difference distribution parameters generated by the VAE algorithm;
and (4) defining the constructed voiceprint model library of the power equipment based on the MSELoss function, and generating an audio abnormal threshold of the normal sound of the transformer.
As a further optimization of the above scheme, the audio anomaly threshold is correspondingly reduced by 1% based on the anomaly sample library update for the transformer.
The invention also discloses a transformer running state monitoring system based on voiceprint, which comprises a computer device, and is characterized in that the computer device is programmed or configured to execute the steps of the transformer running state identification method based on voiceprint image characteristics in any claim, or a storage medium of the computer device is stored with a computer program which is programmed or configured to execute the transformer running state monitoring method based on voiceprint in any claim.
By adopting the technical scheme, compared with the prior art, the transformer running state monitoring method and system based on voiceprint have the following technical effects:
1. according to the invention, the first preset value and the second preset value are set, and the tone quality division is carried out on the original audio data collected by the transformer, so that the transformer can be ensured to collect effective original audio data, and therefore, in the subsequent further analysis process, the interference items can be screened greatly, and the calculated amount in the analysis process is reduced.
2. According to the method, other anomalies which are not stored in the abnormal sample library are marked and used as the updated abnormal sample library sample data, so that the positive feedback of the judgment of the abnormal condition of the target audio data is realized, the capacity of the abnormal sample library is continuously expanded in the analysis process, the comprehensive integration and induction of the target audio data are facilitated, the fault tolerance is enhanced, the original audio collected by the transformer is marked to the maximum extent, and therefore the running state monitoring of the transformer is carried out, and the effect is remarkable.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a schematic flow chart of the classification of target audio data according to the present invention;
fig. 3 is a schematic flow chart of converting original audio data into target audio data according to the present invention.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1 to 3, an embodiment of the present invention discloses a transformer operating state monitoring method based on voiceprint, including:
s100, acquiring original audio data of a transformer;
s200, determining whether the tone quality of the acquired original audio data needs to be corrected or not according to the tone quality of the original audio data acquired by the transformer;
s300, if necessary, performing modification based on the most significant value of the original audio data to obtain valid original audio data.
The method comprises the steps that tone quality division is carried out on original audio data collected by a transformer through setting a first preset value and a second preset value, so that the transformer can be ensured to collect effective original audio data;
specifically, for example, in the natural environment of the transformer, there are sounds emitted by living beings, sounds emitted by the operation of the transformer equipment, sounds generated by faults of the transformer during the operation process, and sounds generated by interaction of the transformer with other objects during the operation process; for various noises in the environment where the transformer is located, it is necessary to perform sound quality division on the acquired original audio data, so that in the subsequent further analysis process, the interference items can be screened greatly, and the calculation amount in the analysis process is reduced.
More specifically, the tone quality related to the invention comprises the loudness and the timbre of the sound, and whether the loudness is in a first preset value interval or a second preset value interval can be determined by obtaining the loudness, so as to facilitate further audio data processing;
through the acquisition of the tone, the object type division can be effectively carried out on the acquired original audio data, the effective tone is screened for further analysis and processing, and the effect is good.
Further, the modifying based on the most significant value of the original audio data includes:
judging whether the tone quality of the original audio data is greater than or equal to a first preset value or not;
if so, determining that the tone quality of the collected original audio data needs to be corrected;
if not, determining that the tone quality of the collected original audio data does not need to be corrected.
Further, judging whether the tone quality of the original audio data is less than or equal to a second preset value;
if so, determining that the tone quality of the collected original audio data needs to be corrected;
if not, determining that the tone quality of the collected original audio data does not need to be corrected.
Specifically, the specific method of the present invention for the original audio data that needs to be modified is as follows:
and aiming at the original audio data with the tone quality larger than or equal to the first preset value, carrying out loudness optimization on the original audio data to enable the original audio data to be corrected to be between the first preset value and the second preset value.
Further, the method also classifies the fault of the original audio data, and specifically includes the following steps:
acquiring target audio data of the transformer based on the original audio data;
if the audio data of the checking target does not exceed the set audio abnormal threshold, marking the current audio as a normal audio;
if the audio data of the checking target exceeds a set audio abnormal threshold, marking the current audio as an abnormal audio, and storing the abnormal audio in an abnormal sample library;
comparing the abnormal audio confirmed by the calibration with an abnormal sample library based on a Zero-shot Learning algorithm:
if the detected abnormal audio type exists in the abnormal sample library, classifying the detected abnormal audio type into the category of the abnormal sample library;
and if the detected abnormal audio type does not exist in the abnormal sample library, classifying the detected abnormal audio type into an unknown abnormal type.
Particularly, the target audio data is to-be-analyzed data processed based on the original audio data, and audio collected by the transformer is converted to generate quantifiable target audio data so as to judge, summarize and sort the sound collected by the transformer;
the abnormal sample library is formed by recording a large number of transformer equipment and sound samples generated by the interaction of the transformer equipment and other objects, and the sound samples are target audio data which are converted based on original audio data;
further, the unknown abnormal category audio data is marked and classified based on an Active Learning algorithm, and the updated abnormal sample library is as follows:
recording the original abnormal sample library as SQL 1, recording the times of any target audio Data to the classes classified as the abnormal sample library or unknown abnormal classes as n, and recording the target audio Data corresponding to the abnormal sample library as Data a1，Data a2，…Data axRecording the target audio Data corresponding to the unknown abnormality as Data b1，Data b2，…Data byWherein n is x + y;
the nth exception sample library is:
SQL n＝SQL 1+Data a1+Data a2+…Data ax+Data b1+…+Data by (1)
the invention also marks the unknown abnormal category of the target audio data, realizes the positive feedback of the abnormal condition judgment of the target audio data by marking other abnormal categories which are not stored in the abnormal sample library and using the other abnormal categories as the updated abnormal sample library sample data, so that the capacity of the abnormal sample library is continuously expanded in the analysis process, the comprehensive integration and summarization of the target audio data are facilitated, the fault tolerance is enhanced, the original audio collected by the transformer is marked to the maximum extent, the running state monitoring of the transformer is carried out, and the effect is obvious.
Particularly, in the process of practical application, for unknown abnormal categories in the sample data of the abnormal sample library, technical means such as expert marking can be used for marking, so that the target audio data automatically marked as the unknown abnormal categories becomes abnormal items manually marked, and subsequent processing is facilitated.
Further, the process of converting the target audio data based on the original audio collected by the transformer specifically includes the following steps:
s400, acquiring original audio data of a transformer environment;
s500, pre-emphasis processing, framing and windowing are carried out on the collected original audio data, and a plurality of pre-processed audios with frames as units are generated;
s600, acquiring a plurality of preprocessed audio frequencies, and acquiring frequency spectrum information corresponding to the preprocessed audio frequencies based on fast Fourier transform;
s700, acquiring Mel frequency spectrums from the plurality of frequency spectrum information by adopting a Mel filter bank;
and S800, performing cepstrum analysis on the Mel spectrum to obtain Mel frequency cepstrum coefficient MFCC.
Further, the cepstrum analysis process specifically includes the following steps:
collecting a plurality of Mel frequency spectrums for logarithmic operation;
implementing an inverse transform by a discrete cosine transform;
and selecting the coefficient after discrete cosine transform as an MFCC coefficient to generate target audio data.
Further, after obtaining the plurality of MFCC coefficients, the generating of the audio anomaly threshold comprises:
compressing a plurality of MFCC coefficients based on an Auto-encoder algorithm, generating a space characterization number and outputting the space characterization number;
based on the VAE algorithm, converting the MFCC coefficients into statistical distribution parameters, namely a mean value and a standard deviation;
constructing a power equipment voiceprint model library based on the spatial representation number output by the Auto-encoder algorithm and the mean value and standard difference distribution parameters generated by the VAE algorithm;
and (4) defining the constructed voiceprint model library of the power equipment based on the MSELoss function, and generating an audio abnormal threshold of the normal sound of the transformer.
Particularly, the invention preferably sets the numerical value corresponding to 15% of data quantity in the voiceprint model library of the electrical equipment as the audio anomaly threshold value.
Further, based on the update of the abnormal sample library of the transformer, the audio frequency abnormal threshold value is correspondingly reduced by 1%.
Particularly, the sample data of the abnormal sample library is continuously updated based on the target audio data, so that the positive feedback of the abnormal condition judgment of the target audio data is realized, the accuracy of the transformer operation condition judgment of the transformer for the transformer-based sound collection is improved, and the audio abnormal threshold is adaptively reduced.
In addition, the system for monitoring the running state of the transformer based on the voiceprint provided by the embodiment of the invention and the method for monitoring the running state of the transformer based on the voiceprint provided by the embodiment of the invention belong to the same concept, and the specific implementation process is described in the method embodiment in detail, which is not described herein again.
The above description is only a preferred embodiment of the application and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention herein disclosed is not limited to the particular combination of features described above, but also encompasses other arrangements formed by any combination of the above features or their equivalents without departing from the inventive concept. For example, the above features may be replaced with (but not limited to) features having similar functions disclosed in the present application.
Claims (10)
1. A transformer running state monitoring method based on voiceprint is characterized by comprising the following steps:
acquiring original audio data of a transformer;
determining whether the tone quality of the acquired original audio data needs to be corrected or not according to the tone quality of the original audio data acquired by the transformer;
if necessary, a correction is made based on the most significant value of the original audio data to obtain valid original audio data.
2. The method for monitoring the operating condition of the transformer based on the voiceprint according to claim 1, wherein the modification based on the maximum value of the original audio data comprises:
judging whether the tone quality of the original audio data is greater than or equal to a first preset value or not;
if so, determining that the tone quality of the collected original audio data needs to be corrected;
if not, determining that the tone quality of the collected original audio data does not need to be corrected.
3. The transformer operating state monitoring method based on the voiceprint according to claim 2, wherein whether the tone quality of the original audio data is smaller than or equal to a second preset value is judged;
if so, determining that the tone quality of the collected original audio data needs to be corrected;
if not, determining that the tone quality of the collected original audio data does not need to be corrected.
4. The method for monitoring the operating state of the transformer based on the voiceprint according to claim 1, wherein the method further classifies the fault of the original audio data, and specifically comprises the following steps:
acquiring target audio data of the transformer based on the original audio data;
if the audio data of the checking target does not exceed the set audio abnormal threshold, marking the current audio as a normal audio;
if the audio data of the checking target exceeds a set audio abnormal threshold, marking the current audio as an abnormal audio, and storing the abnormal audio in an abnormal sample library;
comparing the abnormal audio confirmed by the calibration with an abnormal sample library based on a Zero-shot Learning algorithm:
if the detected abnormal audio type exists in the abnormal sample library, classifying the detected abnormal audio type into the category of the abnormal sample library;
and if the detected abnormal audio type does not exist in the abnormal sample library, classifying the detected abnormal audio type into an unknown abnormal type.
5. The method for monitoring the running state of the transformer based on the voiceprint according to claim 4, wherein the unknown abnormal class audio data is marked and classified based on an Active Learning algorithm, and the updated abnormal sample library is as follows:
recording the original abnormal sample library as SQL 1, recording the times of classifying any target audio Data into the abnormal sample library or unknown abnormal class as n, and recording the target audio Data corresponding to the abnormal sample library as Data a1，Data a2，…Data axRecording the target audio Data corresponding to the unknown abnormality as Data b1，Data b2，…Data byWherein n is x + y;
the nth exception sample library is:
SQL n＝SQL 1+Data a1+Data a2+…Data ax+Data b1+…+Data by (1)。
6. the method for monitoring the operating state of the transformer based on the voiceprint according to claim 4, wherein the process of converting the target audio data based on the original audio collected by the transformer specifically comprises the following steps:
acquiring original audio data of a transformer environment;
pre-emphasis processing, framing and windowing are carried out on the collected original audio data, and a plurality of pre-processed audios with frames as units are generated;
acquiring a plurality of preprocessed audio frequencies, and acquiring frequency spectrum information corresponding to the preprocessed audio frequencies based on fast Fourier transform;
acquiring Mel frequency spectrum from a plurality of frequency spectrum information by adopting a Mel filter bank;
cepstrum analysis is performed on top of the Mel spectrum to obtain Mel-frequency cepstral coefficients MFCC.
7. The method for monitoring the operating state of the transformer based on the voiceprint according to claim 6, wherein the cepstrum analysis process specifically comprises the following steps:
collecting a plurality of Mel frequency spectrums for logarithmic operation;
implementing an inverse transform by a discrete cosine transform;
and selecting the coefficient after discrete cosine transform as an MFCC coefficient to generate target audio data.
8. The method of claim 7, wherein after obtaining the plurality of MFCC coefficients, the generating of the audio anomaly threshold comprises:
compressing a plurality of MFCC coefficients based on an Auto-encoder algorithm to generate and output a space characterization number;
converting a plurality of said MFCC coefficients into statistical distribution parameters-mean and standard deviation based on a VAE algorithm;
constructing a power equipment voiceprint model library based on the spatial representation number output by the Auto-encoder algorithm and the mean value and standard difference distribution parameters generated by the VAE algorithm;
and (4) defining the constructed voiceprint model library of the power equipment based on the MSELoss function, and generating an audio abnormal threshold of the normal sound of the transformer.
9. The method of claim 5, wherein the audio anomaly threshold is correspondingly reduced by 1% based on an anomaly sample library update for the transformer.
10. A transformer operation state monitoring system based on voiceprint, comprising a computer device, wherein the computer device is programmed or configured to execute the steps of the transformer operation state identification method based on voiceprint image characteristics as claimed in any one of claims 1 to 9, or a storage medium of the computer device has stored thereon a computer program programmed or configured to execute the transformer operation state monitoring method based on voiceprint as claimed in any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111467440.4A CN114171058A (en) | 2021-12-03 | 2021-12-03 | Transformer running state monitoring method and system based on voiceprint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111467440.4A CN114171058A (en) | 2021-12-03 | 2021-12-03 | Transformer running state monitoring method and system based on voiceprint |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114171058A true CN114171058A (en) | 2022-03-11 |
Family
ID=80482783
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111467440.4A Pending CN114171058A (en) | 2021-12-03 | 2021-12-03 | Transformer running state monitoring method and system based on voiceprint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114171058A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101729034A (en) * | 2008-10-31 | 2010-06-09 | 美商富迪科技股份有限公司 | Speech processing apparatus, dynamic range control module, and method for amplitude adjustment for a speech signal |
| CN109817241A (en) * | 2019-02-18 | 2019-05-28 | 腾讯音乐娱乐科技（深圳）有限公司 | Audio-frequency processing method, device and storage medium |
| CN111131616A (en) * | 2019-12-28 | 2020-05-08 | 科大讯飞股份有限公司 | Audio sharing method based on intelligent terminal and related device |
| CN111739557A (en) * | 2020-06-19 | 2020-10-02 | 浙江讯飞智能科技有限公司 | Equipment fault positioning method, device, equipment and storage medium |
| CN112763214A (en) * | 2020-12-31 | 2021-05-07 | 南京信息工程大学 | Rolling bearing fault diagnosis method based on multi-label zero-sample learning |
| US20210182618A1 (en) * | 2018-10-29 | 2021-06-17 | Hrl Laboratories, Llc | Process to learn new image classes without labels |
| CN113611331A (en) * | 2021-07-30 | 2021-11-05 | 合肥科大智能机器人技术有限公司 | Transformer voiceprint anomaly detection method |
-
2021
- 2021-12-03 CN CN202111467440.4A patent/CN114171058A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101729034A (en) * | 2008-10-31 | 2010-06-09 | 美商富迪科技股份有限公司 | Speech processing apparatus, dynamic range control module, and method for amplitude adjustment for a speech signal |
| US20210182618A1 (en) * | 2018-10-29 | 2021-06-17 | Hrl Laboratories, Llc | Process to learn new image classes without labels |
| CN109817241A (en) * | 2019-02-18 | 2019-05-28 | 腾讯音乐娱乐科技（深圳）有限公司 | Audio-frequency processing method, device and storage medium |
| CN111131616A (en) * | 2019-12-28 | 2020-05-08 | 科大讯飞股份有限公司 | Audio sharing method based on intelligent terminal and related device |
| CN111739557A (en) * | 2020-06-19 | 2020-10-02 | 浙江讯飞智能科技有限公司 | Equipment fault positioning method, device, equipment and storage medium |
| CN112763214A (en) * | 2020-12-31 | 2021-05-07 | 南京信息工程大学 | Rolling bearing fault diagnosis method based on multi-label zero-sample learning |
| CN113611331A (en) * | 2021-07-30 | 2021-11-05 | 合肥科大智能机器人技术有限公司 | Transformer voiceprint anomaly detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Roy et al. | Autocorrelation aided random forest classifier-based bearing fault detection framework | |
| US11022633B2 (en) | Enhanced system and method for conducting PCA analysis on data signals | |
| CN111259921B (en) | Transformer sound abnormality detection method based on improved wavelet packet and deep learning | |
| CN111562458B (en) | Power transformer fault diagnosis method and power transformer fault diagnosis device | |
| CN113763986B (en) | Abnormal sound detection method for air conditioner indoor unit based on sound classification model | |
| CN116075733A (en) | Battery management system for classifying battery modules | |
| CN112599134A (en) | Transformer sound event detection method based on voiceprint recognition | |
| CN111918196B (en) | Method, device and equipment for diagnosing recording abnormity of audio collector and storage medium | |
| CN113740635A (en) | Electrical equipment fault diagnosis method, terminal and multi-probe sensing device | |
| CN112052712B (en) | Power equipment state monitoring and fault identification method and system | |
| CN113470694A (en) | Remote listening monitoring method, device and system for hydraulic turbine set | |
| CN116564332A (en) | Frequency response analysis method, device, equipment and storage medium | |
| CN115376526A (en) | Power equipment fault detection method and system based on voiceprint recognition | |
| CN111024147A (en) | Component mounting detection method and device based on CNNs, electronic equipment and storage medium | |
| CN117331790A (en) | Machine room fault detection method and device for data center | |
| CN114171058A (en) | Transformer running state monitoring method and system based on voiceprint | |
| CN111860241A (en) | Power equipment discharge fault identification method based on wavelet packet analysis | |
| CN116699400A (en) | Generator rotor short-circuit fault monitoring system, method and readable storage medium | |
| CN117074852A (en) | Power distribution network electric energy monitoring and early warning management method and system | |
| CN114157023B (en) | Distribution transformer early warning information acquisition method | |
| CN114121025A (en) | Voiceprint fault intelligent detection method and device for substation equipment | |
| CN114118208A (en) | Transformer fault judgment method and device based on multivariate information and electronic equipment | |
| CN112114215A (en) | Transformer aging evaluation method and system based on error back propagation algorithm | |
| JPH08320251A (en) | Sound and vibration diagnostic method in equipment | |
| US20060074820A1 (en) | Identifying a state of a data storage drive using an artificial neural network generated model |
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 |