CN110307899A - Sound anomaly detection system based on deep learning - Google Patents
Sound anomaly detection system based on deep learning Download PDFInfo
- Publication number
- CN110307899A CN110307899A CN201910688625.4A CN201910688625A CN110307899A CN 110307899 A CN110307899 A CN 110307899A CN 201910688625 A CN201910688625 A CN 201910688625A CN 110307899 A CN110307899 A CN 110307899A
- Authority
- CN
- China
- Prior art keywords
- sound
- module
- detection module
- output end
- decibel
- 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.)
- Withdrawn
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 66
- 238000013135 deep learning Methods 0.000 title claims abstract description 17
- 238000004891 communication Methods 0.000 claims abstract description 18
- 238000000605 extraction Methods 0.000 claims abstract description 18
- 230000006870 function Effects 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims abstract description 5
- 238000003860 storage Methods 0.000 claims description 24
- 230000005856 abnormality Effects 0.000 claims description 13
- 238000000034 method Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 8
- 235000015170 shellfish Nutrition 0.000 claims description 4
- 230000008054 signal transmission Effects 0.000 claims description 3
- 241000208340 Araliaceae Species 0.000 claims 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 1
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 1
- 235000008434 ginseng Nutrition 0.000 claims 1
- 238000012423 maintenance Methods 0.000 abstract description 4
- 230000002159 abnormal effect Effects 0.000 abstract description 3
- 238000012806 monitoring device Methods 0.000 abstract description 3
- 238000013461 design Methods 0.000 abstract description 2
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 230000000694 effects Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; 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
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; 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
Abstract
The invention discloses a sound anomaly detection system based on deep learning, which comprises a main control module, a sound extraction module, a distance detection module, a timing detection module, a sound decibel detection module, an alarm module and a communication module, the output end of the main control module is respectively connected with the input ends of the sound extraction module, the distance detection module, the timing detection module, the sound decibel detection module, the alarm module and the communication module, the output end of the alarm module is connected with the input end of the communication module, the output end of the communication module is connected with monitoring equipment, the monitoring device is used for transmitting the detected sound abnormal information to the background management end, the sound extraction module comprises a sound acquisition unit, a sound identification unit and a sound processing unit which are sequentially connected, and the input end of the sound extraction module is connected with a microphone. The invention has reasonable design and layout, safety and high efficiency, achieves the functions of warning and quick maintenance, improves the safety performance of the system and is suitable for popularization.
Description
Technical field
The present invention relates to sound detection technical field more particularly to a kind of cacophonia detection systems based on deep learning
System.
Background technique
Acoustic vibration detection, application field machinery, environment, electric power, automobile, space flight, sound generate vibration, monitoring device vibration
Moving noise.For sound since vibration generates, the two has close connection, in field of industrial production, produces since mechanical equipment operates
Raw vibration, the sound of sending are known as industrial noise.It is analyzed according to the signal of sound and vibration signal, judges industrial equipment or product
Quality condition, it has also become one of the important means of field of industry detection.The built-in quality and fortune of industrial noise vibration and equipment
Row situation is closely related, therefore, by the vibration noise of monitoring device, the noise control of working environment both may be implemented, and ensures
Personnel safety, and can analyze the operational situation of mechanical equipment by analyzing noise vibration signal feature, realize mechanical equipment event
Hinder on-line measurement diagnosis, loss caused by avoiding because of equipment fault.
It will appear various sound in equipment running process in many monitoring places, it may be possible to issue under operating status
Normal sound, it is also possible to the abnormal sound that equipment issues in case of a fault.Administrative staff after special training,
The normal or abnormal situation of equipment can be judged according to the alternative sounds that equipment issues.However require administrative staff's whole day 24 hours
It is unpractical for being monitored to all equipment, and intermittent inspection can not in real time, effectively detect the different of each equipment
Reason condition.Therefore, we have proposed a kind of sound abnormality detecting systems based on deep learning for solving the above problems.
Summary of the invention
Technical problems based on background technology, the cacophonia detection based on deep learning that the invention proposes a kind of
System.
A kind of sound abnormality detecting system based on deep learning proposed by the present invention, including main control module, sound extract
Module, apart from detection module, timing detection module, sound decibel detection module, alarm module and communication module, the master control mould
The output end of block respectively with the sound extraction module, apart from detection module, timing detection module, sound decibel detection module,
Alarm module is connected with the input terminal of communication module, and the input terminal of the output end of the alarm module and the communication module connects
It connects, the output end of the communication module is connected with supervision equipment, and the cacophonia information for will test is transmitted to backstage and manages
The supervision equipment at end is managed, the sound extraction module includes sequentially connected sound collection unit, acoustic recognition unit and sound
Processing unit, wherein the input terminal of sound extraction module is connected with microphone, for collected sound to be passed through identification, processing
Afterwards, voice signal is converted into electric signal transmission to the sound decibel detection module and carries out cacophonia detection.
It preferably, further include information storage module, the information storage module includes that decibel value presets storage unit, timing
Information memory cell and object of reference gauge length storage unit.
Preferably, the output end that the decibel value presets storage unit is connect with the input terminal of sound decibel detection module,
For starting warning device according to the type of detecting instrument and the setting of preset multiple groups decibel value in sound decibel detection process
Sound decibel value.
Preferably, the output end of the clocking information storage unit is connect with the input terminal of timing detection module, is used for root
Start the function that timing detects cacophonia according to the preset period.
Preferably, the output end of the object of reference gauge length storage unit is connect with the input terminal apart from detection module, is used for
In sound decibel detection process real-time measurement detecting instrument between object of reference at a distance from, and according to preset object of reference gauge length
Numerical value and above-mentioned distance versus, obtain sound decibel testing result.
Preferably, the supervision equipment is computer or mobile phone, passes through APP software, information, mail or Advise By Wire.
In the present invention, a kind of sound abnormality detecting system based on deep learning passes through sound extraction module, distance
The setting of detection module, timing detection module, sound decibel detection module and information storage module, to the sound after extraction right
Three models detection is carried out in the case of target, it can be to the detecting instrument under different conditions by referring to object gauge length harmony cent shellfish
The mode of detection double combination is managed, and the mode for cooperating timing to detect carries out cacophonia detection, reaches and accurately detects
As a result, cacophonia information is transmitted to back-stage management end using communication module simultaneously, to prompt the working condition of detection device
Exception improve the security performance of system and then convenient for warning and the effect of Fast-Maintenance.Design layout of the present invention is reasonable, peace
Overall height effect, has the function that warning and Fast-Maintenance, improves the security performance of system, is suitable for promoting.
Detailed description of the invention
Fig. 1 is a kind of functional block diagram of the sound abnormality detecting system based on deep learning proposed by the present invention；
Fig. 2 is a kind of flow diagram of the sound abnormality detecting system based on deep learning proposed by the present invention.
Specific embodiment
Combined with specific embodiments below the present invention is made further to explain.
Embodiment
Referring to Fig.1-2, a kind of sound abnormality detecting system based on deep learning, including main control module, sound extract mould
Block, apart from detection module, timing detection module, sound decibel detection module, alarm module and communication module, the main control module
Output end respectively with the sound extraction module, apart from detection module, timing detection module, sound decibel detection module, report
Alert module is connected with the input terminal of communication module, and the output end of the alarm module is connect with the input terminal of the communication module,
The output end of the communication module is connected with supervision equipment, and the cacophonia information for will test is transmitted to back-stage management end
Supervision equipment, the sound extraction module includes sequentially connected sound collection unit, acoustic recognition unit and acoustic processing
Unit, wherein the input terminal of sound extraction module is connected with microphone, for by collected sound after identifying, handling,
Voice signal is converted into electric signal transmission to the sound decibel detection module and carries out cacophonia detection.
Further include information storage module in the present embodiment, the information storage module include decibel value preset storage unit,
Clocking information storage unit and object of reference gauge length storage unit, the decibel value preset the output end and sound decibel of storage unit
The input terminal of detection module connects, and is used in sound decibel detection process, according to the type of detecting instrument and preset multiple groups
The sound decibel value of decibel value setting starting warning device, the output end and timing detection module of the clocking information storage unit
Input terminal connection, for started according to the preset period timing detect cacophonia function, the object of reference gauge length
The output end of storage unit is connect with the input terminal apart from detection module, for the real-time measurement inspection in sound decibel detection process
Survey instrument between object of reference at a distance from, and according to the numerical value of preset object of reference gauge length and above-mentioned distance versus, obtain sound and divide
Shellfish testing result, the supervision equipment are computer or mobile phone, pass through APP software, information, mail or Advise By Wire.
In the present embodiment, mould is detected by sound extraction module, apart from detection module, timing detection module, sound decibel
The setting of block and information storage module carries out Three models detection to the sound after extraction to target, can be to not
It is managed with the detecting instrument under state by referring to the mode of object gauge length harmony cent shellfish detection double combination, and cooperates timing
The mode of detection carries out cacophonia detection, reaches accurately testing result, while utilizing communication module by cacophonia information
It is transmitted to back-stage management end, to prompt the exception of the working condition of detection device, and then convenient for warning the effect with Fast-Maintenance,
The security performance of raising system.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
Anyone skilled in the art in the technical scope disclosed by the present invention, according to the technique and scheme of the present invention and its
Inventive concept is subject to equivalent substitution or change, should be covered by the protection scope of the present invention.
Claims (6)
1. a kind of sound abnormality detecting system based on deep learning, including main control module, sound extraction module, distance detection mould
Block, timing detection module, sound decibel detection module, alarm module and communication module, which is characterized in that the main control module
Output end is respectively with the sound extraction module, apart from detection module, timing detection module, sound decibel detection module, alarm
Module is connected with the input terminal of communication module, and the output end of the alarm module is connect with the input terminal of the communication module, institute
The output end for stating communication module is connected with supervision equipment, and the cacophonia information for will test is transmitted to back-stage management end
Supervision equipment, the sound extraction module include sequentially connected sound collection unit, acoustic recognition unit and acoustic processing list
Member, wherein the input terminal of sound extraction module is connected with microphone, for after identifying, handling, inciting somebody to action collected sound
Voice signal is converted to electric signal transmission to the sound decibel detection module and carries out cacophonia detection.
2. a kind of sound abnormality detecting system based on deep learning according to claim 1, which is characterized in that further include
Information storage module, the information storage module include that decibel value presets storage unit, clocking information storage unit and object of reference
Gauge length storage unit.
3. a kind of sound abnormality detecting system based on deep learning according to claim 2, which is characterized in that described point
The output end that shellfish value presets storage unit is connect with the input terminal of sound decibel detection module, in sound decibel detection process
In, according to the sound decibel value of the type of detecting instrument and preset multiple groups decibel value setting starting warning device.
4. a kind of sound abnormality detecting system based on deep learning according to claim 2, which is characterized in that the meter
When information memory cell output end connect with the input terminal of timing detection module, it is fixed for being started according to the preset period
When detect cacophonia function.
5. a kind of sound abnormality detecting system based on deep learning according to claim 2, which is characterized in that the ginseng
It is connect according to the output end of object gauge length storage unit with the input terminal apart from detection module, for real in sound decibel detection process
When measuring and testing instrument between object of reference at a distance from, and according to the numerical value of preset object of reference gauge length and above-mentioned distance versus, obtain
To sound decibel testing result.
6. a kind of sound abnormality detecting system based on deep learning according to claim 1, which is characterized in that the prison
Equipment is regarded as computer or mobile phone, passes through APP software, information, mail or Advise By Wire.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910688625.4A CN110307899A (en) | 2019-07-29 | 2019-07-29 | Sound anomaly detection system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910688625.4A CN110307899A (en) | 2019-07-29 | 2019-07-29 | Sound anomaly detection system based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110307899A true CN110307899A (en) | 2019-10-08 |
Family
ID=68082344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910688625.4A Withdrawn CN110307899A (en) | 2019-07-29 | 2019-07-29 | Sound anomaly detection system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110307899A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021217431A1 (en) * | 2020-04-28 | 2021-11-04 | 深圳市大疆创新科技有限公司 | Noise reduction method, state determination method, and electronic device |
-
2019
- 2019-07-29 CN CN201910688625.4A patent/CN110307899A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021217431A1 (en) * | 2020-04-28 | 2021-11-04 | 深圳市大疆创新科技有限公司 | Noise reduction method, state determination method, and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110647133B (en) | Rail transit equipment state detection maintenance method and system | |
| EP3085937B1 (en) | System and method for detecting vehicle system faults | |
| CN110275094B (en) | High-voltage cable partial discharge acousto-optic combined detection method and system | |
| CN107345857A (en) | A kind of electro spindle condition monitoring and failure diagnosis system and its monitoring, diagnosing method | |
| CN107144303A (en) | A kind of fault detection system of mineral electromechanical equipment, method and device | |
| CN104090252B (en) | Comprehensive detecting system and method for cable type fault indicator | |
| CN111336100A (en) | Water pump fault diagnosis system | |
| CN102183697A (en) | System for monitoring noise and vibration of power transformer | |
| KR20090010430A (en) | Apparatus for detecting mechanical trouble | |
| CN113467433A (en) | Method and device for detecting faults of mechanical equipment | |
| JP2019067197A (en) | Method for detecting trouble sign | |
| US20210149387A1 (en) | Facility failure prediction system and method for using acoustic signal of ultrasonic band | |
| KR102545672B1 (en) | Method and apparatus for machine fault diagnosis | |
| CN116383633A (en) | Method and system for detecting faults of machine-made sand vibrating screen through multi-factor comprehensive analysis | |
| CN103821750B (en) | A kind of ventilator stall based on electric current and surge monitoring and diagnostic method | |
| CN110307899A (en) | Sound anomaly detection system based on deep learning | |
| CN113483815A (en) | Mechanical fault monitoring system based on industrial big data | |
| CN105784364A (en) | Bearing fault diagnosis method based on total experience mode decomposition and fractal box dimensions | |
| CN108444592A (en) | Wireless vibration monitoring and fault diagnosis system | |
| CN112945531A (en) | Method, device and equipment for detecting cracks of fan blade and computer storage medium | |
| CN203855261U (en) | Real-time monitoring system for elevator running | |
| CN110057587A (en) | A kind of nuclear power pump bearing intelligent failure diagnosis method and system | |
| KR101248232B1 (en) | Diagnosis method and system on deteriration of parallel driven utility equipments | |
| CN104165683A (en) | Online power capacitor vibration monitoring device | |
| CN201843767U (en) | Failure audio frequency analysis and diagnosis device for turbine type compressor |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20191008 |