CN117373478A - Voiceprint recognition system for transformer - Google Patents
Voiceprint recognition system for transformer Download PDFInfo
- Publication number
- CN117373478A CN117373478A CN202311384754.7A CN202311384754A CN117373478A CN 117373478 A CN117373478 A CN 117373478A CN 202311384754 A CN202311384754 A CN 202311384754A CN 117373478 A CN117373478 A CN 117373478A
- Authority
- CN
- China
- Prior art keywords
- voiceprint
- transformer
- data
- module
- feature extraction
- 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
- 230000002159 abnormal effect Effects 0.000 claims abstract description 34
- 238000000605 extraction Methods 0.000 claims abstract description 28
- 238000012545 processing Methods 0.000 claims abstract description 21
- 238000012544 monitoring process Methods 0.000 claims abstract description 9
- 238000012423 maintenance Methods 0.000 claims abstract description 7
- 230000005856 abnormality Effects 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims abstract description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 13
- 238000003062 neural network model Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000007619 statistical method Methods 0.000 claims description 4
- 238000000691 measurement method Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 5
- 230000002035 prolonged effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
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
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
-
- 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
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention discloses a voiceprint recognition system for a transformer, relates to the technical field of artificial intelligent voiceprint recognition, and solves the technical problems of inaccurate fault detection and untimely maintenance of the transformer caused by less management staff and insufficient experience on the existing power network line; the method comprises the following steps: the voiceprint acquisition module acquires voiceprint data of the transformer in real time; the data processing module is used for processing the collected transformer voiceprint data; the voiceprint feature extraction module is used for extracting and analyzing voiceprint features of the preprocessed data; if the similarity of the extracted voiceprint features is smaller than a similarity threshold, displaying abnormality and sending the abnormal voiceprint features to a voiceprint recognition module; the voiceprint recognition module is used for recognizing abnormal categories of the transformer; the cloud application platform is used for monitoring the running state of the transformer in real time and early warning the hidden trouble of the fault; the functions of accurate prediction and real-time early warning of faults are realized.
Description
Technical Field
The invention belongs to the field of transformer detection, relates to an artificial intelligent voiceprint recognition technology, and particularly relates to a voiceprint recognition system for a transformer.
Background
The operation fault of the power transformer is a key cause of large-area power failure of the system, and the development of intelligent operation and detection is an effective means for guaranteeing the safe operation of the power transformer.
Nowadays, management staff on a power network line are fewer and fewer, and due to insufficient technical experience, a product hung on the network is never overhauled, normal and abnormal operation is not known, and fully non-automatic equipment is processed according to full-automatic equipment, so that a more accurate and intelligent fault identification and maintenance mode is needed in the whole market.
The present invention therefore proposes a voiceprint recognition system for a transformer.
Disclosure of Invention
The purpose of the application is to provide a voiceprint recognition system for transformer, solved the manager on the present power network line and little and experience is not enough, leads to the transformer fault detection inaccurate and the untimely problem of maintenance.
To achieve the above object, the present application provides a voiceprint recognition system for a transformer, including: the system comprises a voiceprint acquisition module, a data processing module, a voiceprint feature extraction module, a voiceprint identification module and a cloud application platform;
the voiceprint acquisition module is used for acquiring the voiceprint data of the transformer in real time and sending the voiceprint data to the voiceprint feature extraction module;
the data processing module is used for processing the collected transformer voiceprint data and sending the collected transformer voiceprint data to the data processing module;
the voiceprint feature extraction module is used for extracting and analyzing voiceprint features of the preprocessed data;
if the similarity of the extracted voiceprint features is smaller than a similarity threshold, displaying abnormality and sending the abnormal voiceprint features to a voiceprint recognition module;
the voiceprint recognition module is used for recognizing the abnormal type of the transformer and sending the abnormal type to the cloud application platform;
the cloud application platform is used for monitoring the running state of the transformer in real time and early warning the hidden trouble.
Further, the voiceprint acquisition device further comprises a position acquisition module, wherein the position acquisition module is used for acquiring position information of the transformer and sending the position information to the cloud application platform.
Further, the data processing module is used for processing the collected transformer voiceprint data:
preprocessing transformer voiceprint data at intervals of T time, extracting a Mel spectrogram of each section of voiceprint data, converting the Mel spectrogram into RGB images, and sending the RGB images to a voiceprint feature extraction module as preprocessed voiceprint data.
Further, the voiceprint feature extraction module is configured to perform voiceprint feature extraction and analysis on the preprocessed data, and includes:
the voiceprint feature extraction module stores normal voiceprint feature library, matches the extracted voiceprint features with the normal voiceprint feature library, judges the similarity degree between the voiceprint features through a similarity measurement method, and sets a similarity threshold.
Further, the voiceprint feature extraction module is configured to perform voiceprint feature extraction and analysis on the preprocessed data, and further includes:
if the extracted voiceprint feature similarity is greater than or equal to the similarity threshold, displaying normal, and sending a normal signal to the cloud application platform.
Further, the voiceprint recognition module is built based on a CNN neural network model, and comprises the following steps of;
step S1: the voiceprint recognition module collects abnormal voiceprint sample data in advance, performs preprocessing and feature extraction, and divides the processed data into a training data set and a verification data set;
labeling each abnormal voiceprint sample with a corresponding abnormal category label;
step S2: inputting the preprocessed voiceprint sample and the corresponding label into a CNN neural network for training;
step S3: when the prediction accuracy reaches 95% or more, stopping training, and deploying the trained CNN neural network model into an actual environment for real-time abnormal voiceprint recognition.
Further, when a new abnormal voiceprint sample appears, the new abnormal voiceprint sample is input into the CNN neural network model, and the model library is trained, iterated and optimized and updated for the existing model.
Further, the cloud application platform comprises a user homepage unit, a voiceprint sample library management unit, an algorithm model library management unit, a maintenance management unit and a statistical analysis unit.
Further, the user homepage unit displays the position of the transformer substation at the regional end where the user is located, the monitoring result of the transformer substation voiceprint and the running state of the voiceprint acquisition device, and immediately sends corresponding staff to repair the transformer with faults.
Compared with the prior art, the invention has the beneficial effects that:
the voice print acquisition module is used for acquiring voice print data of the transformer in real time, the data processing module is used for processing the acquired voice print data of the transformer, the voice print feature extraction module is used for extracting and analyzing voice print features of the preprocessed data, the voice print recognition module is used for recognizing abnormal types of the transformer, the cloud application platform monitors the running state of the transformer in real time, and early warning is carried out on hidden trouble occurrence; abnormal operation state of the transformer is detected through an artificial intelligent voiceprint recognition technology, manual judgment inaccuracy is obviously improved, meanwhile, labor cost can be reduced to realize intelligent management, and service life of equipment is effectively prolonged.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required in the prior art and the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a block diagram of a voiceprint recognition system for a transformer according to the present invention.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
Referring to fig. 1 specifically, a voiceprint recognition system for a transformer includes a voiceprint acquisition module, a data processing module, a voiceprint feature extraction module, a voiceprint recognition module, and a cloud application platform;
the voiceprint acquisition module is used for acquiring the voiceprint data of the transformer in real time and sending the voiceprint data to the data processing module;
in one embodiment, the voiceprint data of the transformer is collected by using a voiceprint collecting device, and the installation position of the voiceprint collecting device is set according to actual experience;
the voiceprint acquisition device further comprises a position acquisition module, wherein the position acquisition module is used for acquiring position information of the transformer and sending the position information to the cloud application platform;
the voiceprint acquisition device can be a monitoring voiceprint acquisition device or a patch voiceprint acquisition device, wherein the monitoring voiceprint acquisition device can identify a 20-20KHz audio range and is vertically installed or hoisted; the patch voiceprint acquisition equipment adopts self-adaptive dynamic noise reduction treatment to eliminate the difference of sound at a far distance and a near distance, so that the sound output amplitude is more balanced, the sound playback effect is smoother and more comfortable, the equipment is adsorbed on a transformer base in a magnetic attraction mode, and the voiceprint information of the transformer is monitored at a short distance;
the data processing module is used for processing the collected transformer voiceprint data, preprocessing the transformer voiceprint data every T time, extracting a Mel spectrogram of each section of voiceprint data, converting the Mel spectrogram into an RGB image, and sending the RGB image to the voiceprint feature extraction module as preprocessed voiceprint data;
the voiceprint feature extraction module is used for extracting and analyzing voiceprint features of the preprocessed data;
the voiceprint feature extraction module stores normal voiceprint feature library, matches the extracted voiceprint features with the normal voiceprint feature library, judges the similarity degree between the voiceprint features through a similarity measurement method such as Euclidean distance, cosine similarity and the like, and sets a similarity threshold;
if the similarity of the extracted voiceprint features is smaller than a similarity threshold, displaying abnormality and sending the abnormal voiceprint features to a voiceprint recognition module;
if the extracted voiceprint feature similarity is greater than or equal to a similarity threshold, displaying normal, and sending a normal signal to the cloud application platform;
the voiceprint recognition module is used for recognizing the abnormal type of the transformer and sending the abnormal type to the cloud application platform;
specifically, the voiceprint recognition module is built based on a CNN neural network model, and comprises the following steps of;
step S1: the voiceprint recognition module collects abnormal voiceprint sample data in advance, performs preprocessing and feature extraction, and divides the processed data into a training data set and a verification data set;
each abnormal voiceprint sample is marked with a corresponding abnormal category label, for example: over-current, overload, DC magnetic bias, short circuit impact, component looseness and other types of anomalies;
step S2: inputting the preprocessed voiceprint sample and the corresponding label into a CNN neural network for training, and continuously adjusting the weight of the CNN neural network to enable the predicted result of the model to be as close as possible to the real result;
step S3: when the prediction accuracy reaches 95% or more, stopping training, and deploying the trained CNN neural network model into an actual environment for real-time abnormal voiceprint recognition;
when a new abnormal voiceprint sample appears, the new abnormal voiceprint sample is input into a CNN neural network model, and the existing model is trained, iterated and optimally updated to a model library;
the cloud application platform is used for monitoring the running state of the transformer in real time and early warning the hidden trouble of the fault;
the cloud application platform comprises a user homepage unit, a voiceprint sample library management unit, an algorithm model library management unit, a maintenance management unit and a statistical analysis unit;
specifically, a user logs in a cloud application platform and jumps to a user homepage unit according to user account authority, the user homepage unit displays the position of a transformer substation at the regional end where the user is located, the voiceprint monitoring result of the transformer substation, the running state of a voiceprint acquisition device and the like, and immediately sends corresponding staff to repair a transformer with faults;
the voiceprint sample library management unit: the method realizes uploading, downloading, browsing, acquisition management, label management and the like of the sample information metadata;
the algorithm model library management unit: the management of the version, detection, deployment state and other information of the voiceprint algorithm model is realized;
the maintenance management unit: providing exception handling state tracking management, providing closed loop management of exception-derived patrol events;
the statistical analysis unit: providing a comprehensive data analysis management function, and realizing the management and control of equipment running state, history information and abnormal data tracking;
the voice print acquisition module is used for acquiring voice print data of the transformer in real time, the data processing module is used for processing the acquired voice print data of the transformer, the voice print feature extraction module is used for extracting and analyzing voice print features of the preprocessed data, the voice print recognition module is used for recognizing abnormal types of the transformer, the cloud application platform monitors the running state of the transformer in real time, and early warning is carried out on hidden trouble occurrence; abnormal operation state of the transformer is detected through an artificial intelligent voiceprint recognition technology, manual judgment inaccuracy is obviously improved, meanwhile, labor cost can be reduced to realize intelligent management, and service life of equipment is effectively prolonged.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas which are obtained by acquiring a large amount of data and performing software simulation to obtain the closest actual situation, and preset parameters and preset thresholds in the formulas are set by a person skilled in the art according to the actual situation or are obtained by simulating a large amount of data.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (9)
1. A voiceprint recognition system for a transformer, comprising: the system comprises a voiceprint acquisition module, a data processing module, a voiceprint feature extraction module, a voiceprint identification module and a cloud application platform;
the voiceprint acquisition module is used for acquiring the voiceprint data of the transformer in real time and sending the voiceprint data to the voiceprint feature extraction module;
the data processing module is used for processing the collected transformer voiceprint data and sending the collected transformer voiceprint data to the data processing module;
the voiceprint feature extraction module is used for extracting and analyzing voiceprint features of the preprocessed data;
if the similarity of the extracted voiceprint features is smaller than a similarity threshold, displaying abnormality and sending the abnormal voiceprint features to a voiceprint recognition module;
the voiceprint recognition module is used for recognizing the abnormal type of the transformer and sending the abnormal type to the cloud application platform;
the cloud application platform is used for monitoring the running state of the transformer in real time and early warning the hidden trouble.
2. The voiceprint recognition system for a transformer of claim 1, wherein the voiceprint acquisition device further comprises a location acquisition module, and the location acquisition module is configured to acquire location information of the transformer and send the location information to the cloud application platform.
3. The voiceprint recognition system for a transformer of claim 1, wherein the data processing module is configured to process collected transformer voiceprint data:
preprocessing transformer voiceprint data at intervals of T time, extracting a Mel spectrogram of each section of voiceprint data, converting the Mel spectrogram into RGB images, and sending the RGB images to a voiceprint feature extraction module as preprocessed voiceprint data.
4. The voiceprint recognition system for a transformer of claim 1, wherein the voiceprint feature extraction module is configured to perform voiceprint feature extraction and analysis on the preprocessed data, comprising:
the voiceprint feature extraction module stores normal voiceprint feature library, matches the extracted voiceprint features with the normal voiceprint feature library, judges the similarity degree between the voiceprint features through a similarity measurement method, and sets a similarity threshold.
5. The voiceprint recognition system for a transformer of claim 1, wherein the voiceprint feature extraction module is configured to perform voiceprint feature extraction and analysis on the preprocessed data, and further comprising:
if the extracted voiceprint feature similarity is greater than or equal to the similarity threshold, displaying normal, and sending a normal signal to the cloud application platform.
6. The voiceprint recognition system for a transformer of claim 1, wherein the voiceprint recognition module is built based on a CNN neural network model, comprising the steps of;
step S1: the voiceprint recognition module collects abnormal voiceprint sample data in advance, performs preprocessing and feature extraction, and divides the processed data into a training data set and a verification data set;
labeling each abnormal voiceprint sample with a corresponding abnormal category label;
step S2: inputting the preprocessed voiceprint sample and the corresponding label into a CNN neural network for training;
step S3: when the prediction accuracy reaches 95% or more, stopping training, and deploying the trained CNN neural network model into an actual environment for real-time abnormal voiceprint recognition.
7. A voiceprint recognition system for a transformer according to claim 1 wherein when a new abnormal voiceprint sample is present, it is input into a CNN neural network model, the existing model is trained, iterated, and the model library is optimally updated.
8. The voiceprint recognition system for a transformer of claim 1, wherein the cloud application platform comprises a user homepage unit, a voiceprint sample library management unit, an algorithm model library management unit, a maintenance management unit, and a statistical analysis unit.
9. The voiceprint recognition system for a transformer of claim 8, wherein the user homepage unit displays the location of the substation at the end of the area where the user is located, the voiceprint monitoring result of the substation, and the operation state of the voiceprint acquisition device, and immediately sends corresponding staff to repair the failed transformer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311384754.7A CN117373478A (en) | 2023-10-23 | 2023-10-23 | Voiceprint recognition system for transformer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311384754.7A CN117373478A (en) | 2023-10-23 | 2023-10-23 | Voiceprint recognition system for transformer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117373478A true CN117373478A (en) | 2024-01-09 |
Family
ID=89396132
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311384754.7A Pending CN117373478A (en) | 2023-10-23 | 2023-10-23 | Voiceprint recognition system for transformer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117373478A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117894317A (en) * | 2024-03-14 | 2024-04-16 | 沈阳智帮电气设备有限公司 | Box-type transformer on-line monitoring method and system based on voiceprint analysis |
-
2023
- 2023-10-23 CN CN202311384754.7A patent/CN117373478A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117894317A (en) * | 2024-03-14 | 2024-04-16 | 沈阳智帮电气设备有限公司 | Box-type transformer on-line monitoring method and system based on voiceprint analysis |
| CN117894317B (en) * | 2024-03-14 | 2024-05-24 | 沈阳智帮电气设备有限公司 | Box-type transformer on-line monitoring method and system based on voiceprint analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110826538B (en) | Abnormal off-duty identification system for electric power business hall | |
| CN110366031B (en) | Vision-based abnormal state monitoring and fault diagnosis method for MES (manufacturing execution system) of digital workshop | |
| CN104091070B (en) | Rail transit fault diagnosis method and system based on time series analysis | |
| CN114063507B (en) | Remote equipment control system based on intelligent agriculture and control method thereof | |
| CN116320832B (en) | Monitoring equipment fault monitoring method and device | |
| CN117373478A (en) | Voiceprint recognition system for transformer | |
| CN113296427B (en) | Building site safety monitoring system based on thing networking | |
| CN112925279A (en) | Fault comprehensive analysis system based on MES system | |
| CN113982850B (en) | Fan comprehensive health analysis method and system integrating high-low frequency signals | |
| CN108776452B (en) | Special equipment field maintenance monitoring method and system | |
| CN110349048B (en) | Substation multidimensional data operation interactive control platform and fault handling method | |
| CN107844067A (en) | A kind of gate of hydropower station on-line condition monitoring control method and monitoring system | |
| CN116142913A (en) | Equipment health state analysis method and system based on big data | |
| CN114244866A (en) | Production equipment supervisory systems based on thing networking | |
| CN114781476A (en) | Fault analysis system and method for measuring equipment | |
| CN116880309A (en) | Factory environment monitoring management system and method | |
| CN109051104A (en) | A kind of bottle placer operating status intelligent Detection and method | |
| CN116383633A (en) | Method and system for detecting faults of machine-made sand vibrating screen through multi-factor comprehensive analysis | |
| CN116244765A (en) | Equipment maintenance management method based on industrial Internet | |
| CN117474357A (en) | Power distribution room operation and maintenance management method and system based on deep learning | |
| CN117032055A (en) | Industrial equipment intelligent control system | |
| CN117129796A (en) | Power grid fault identification system based on big data | |
| CN115356970A (en) | Dynamic loop monitoring system and method based on real-time data and operation and maintenance data | |
| CN115328986A (en) | Power plant safety early warning data analysis processing method and system | |
| CN111679926A (en) | Memory overflow detection method and system |
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 |