CN109738058A - A kind of self study vibrating failure diagnosis method - Google Patents
A kind of self study vibrating failure diagnosis method Download PDFInfo
- Publication number
- CN109738058A CN109738058A CN201910116622.3A CN201910116622A CN109738058A CN 109738058 A CN109738058 A CN 109738058A CN 201910116622 A CN201910116622 A CN 201910116622A CN 109738058 A CN109738058 A CN 109738058A
- Authority
- CN
- China
- Prior art keywords
- failure diagnosis
- self study
- model
- vibration
- signal
- 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
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The present invention relates to a kind of self study vibrating failure diagnosis methods, comprising: (1) acquires vibration signal in real time；(2) signal is AD converted；(3) Fourier transformation is carried out to signal；It (4) is state that any data are all failure by self study vibrating failure diagnosis model initialization, then in the state of determining that self study vibrating failure diagnosis model is normal, using by the frequency values in the vibration signal after Fourier transformation as the first input value in neural network, using vibration amplitude corresponding in frequency range as the second input value in neural network, by the first input value, in second input value input model, the output threshold range being calculated is replaced into the initial threshold range that initialization data in self study vibrating failure diagnosis model is all failure, establish diagnostic model；(5) it keeps diagnostic model no longer to change, as input signal and is inputted in the diagnostic model by the vibration signal that acquires in real time, thus through judging show whether Oscillation Amplitude is normal.
Description
Technical field
The present invention relates to the technical fields of vibrating failure diagnosis, and in particular to a kind of self study vibrating failure diagnosis method.
Background technique
Vibration is caused physical phenomenon during mechanical equipment and electric equipment operation, intuitively shows as the past of object
Multiple movement and vibration noise two parts.For specified equipment under the operating condition of standard, the not clear mathematical model of vibrational waveform,
And waveform is relatively fixed, when there are certain failures, vibrational waveform can change, and the side of model is established by self study process
Whether method, can be for judging vibrational waveform in normal range of operation using the model of foundation.
Summary of the invention
In order to overcome the deficiencies in the prior art and defect, the present invention provides a kind of self study vibrating failure diagnosis
Method.
The technical solution adopted by the present invention to solve the technical problems is: a kind of self study vibrating failure diagnosis method,
It is characterized in that, comprising the following steps:
(1) vibration signal is acquired in real time；
(2) by vibration signal by acquiring after AD conversion into self study vibrating failure diagnosis model；
(3) conversion of the time domain to frequency domain will be realized by Fourier transformation by the vibration signal after AD conversion, obtains vibration letter
Number vibration amplitude corresponding on a different frequency；
(4) it is state that any data are all failure by self study vibrating failure diagnosis model initialization, then learns by oneself determining
Practise vibrating failure diagnosis model it is normal in the state of, using by the frequency values in the vibration signal after Fourier transformation as neural
The first input value in network, it is defeated by first using vibration amplitude corresponding in frequency range as the second input value in neural network
Enter value, in the second input value input model, the output threshold range being calculated is replaced into self study vibrating failure diagnosis model
Middle initialization data is all the initial threshold range of failure, establishes diagnostic model；
(5) it keeps diagnostic model no longer to change, which as input signal and is inputted by the vibration signal that acquires in real time
In type, thus show whether Oscillation Amplitude is normal through judgement.
Further, the vibration signal in the step (1) includes times in Oscillation Amplitude signal or vibration audible signal
It anticipates one kind.
Further, it in the step (4), is determined at self study vibrating failure diagnosis model by artificial or detecting instrument
In the state of normal.
Further, vibration amplitude corresponding in frequency range is re-used as mind after normalized in the step (4)
Through the second input value in network.
Further, in the step (4) the first input value, the second input value correspondingly by pairs of input model
In.
Further, in the step (5), which as input signal and is inputted by the vibration signal that acquires in real time
In model, when the result that diagnostic model is calculated is located in output threshold range, it is judged as that Oscillation Amplitude is normal；Work as diagnosis
When the result that model is calculated is beyond output threshold range, it is judged as that Oscillation Amplitude is improper.
The beneficial effects of the invention are as follows；
(1) by learning collected vibration signal feeding neural network, and by by self study vibrating failure diagnosis
Model initialization is the state that any data are all failure, will pass through frequency values in the vibration signal after Fourier transformation and right
In the pairs of input model of the vibration amplitude answered, the output threshold range being calculated is replaced into self study vibrating failure diagnosis mould
Initialization data is all the initial threshold range of failure in type, to establish diagnostic model, and the vibration by acquiring in real time is believed
Number as input signal and inputting in the diagnostic model, thus show whether Oscillation Amplitude is normal through judgement；This method is not directed to
Particular device does not need to set specific mathematical model for equipment, the Working mould of equipment can be established by self study process
Type is sent into model by acquisition live signal and determines whether equipment is in normal operating conditions, and versatile, practicability is high.
Detailed description of the invention
Fig. 1 is a kind of step flow chart of self study vibrating failure diagnosis method of the present invention.
Specific embodiment
In conjunction with the accompanying drawings, the present invention is further explained in detail.These attached drawings are simplified schematic diagram, only with
Illustration illustrates basic structure of the invention, therefore it only shows the composition relevant to the invention.
The technical solution that the present invention is taken by solution its technical problem are as follows:
As shown in Figure 1, a kind of self study vibrating failure diagnosis method, comprising the following steps:
(1) vibration signal is acquired in real time；
(2) by vibration signal by acquiring after AD conversion into self study vibrating failure diagnosis model；
(3) conversion of the time domain to frequency domain will be realized by Fourier transformation by the vibration signal after AD conversion, obtains vibration letter
Number vibration amplitude corresponding on a different frequency；
(4) it is state that any data are all failure by self study vibrating failure diagnosis model initialization, then learns by oneself determining
Practise vibrating failure diagnosis model it is normal in the state of, using by the frequency values in the vibration signal after Fourier transformation as neural
The first input value in network, it is defeated by first using vibration amplitude corresponding in frequency range as the second input value in neural network
Enter value, in the second input value input model, the output threshold range being calculated is replaced into self study vibrating failure diagnosis model
Middle initialization data is all the initial threshold range of failure, establishes diagnostic model；
(5) it keeps diagnostic model no longer to change, which as input signal and is inputted by the vibration signal that acquires in real time
In type, thus show whether Oscillation Amplitude is normal through judgement.
Specifically, the vibration signal in step (1) include Oscillation Amplitude signal or vibration audible signal in any one,
So as to be inputted according to the different signal that needs to acquire actually judged.
Specifically, in step (4), determine that self study vibrating failure diagnosis model is in normal by artificial or detecting instrument
In the state of, to guarantee that self study vibrating failure diagnosis model is working properly to improve judging result accuracy.
Specifically, vibration amplitude corresponding in frequency range is re-used as neural network after normalized in step (4)
In the second input value, consequently facilitating data statistics and calculating.
Specifically, in step (4) the first input value, the second input value correspondingly by pairs of input model, thus
Guarantee the accuracy of study replacement result.
Specifically, in step (5), which as input signal and is inputted by the vibration signal that acquires in real time
In, when the result that diagnostic model is calculated is located in output threshold range, it is judged as that Oscillation Amplitude is normal；Work as diagnostic model
When the result being calculated is beyond output threshold range, it is judged as that Oscillation Amplitude is improper.
Taking the above-mentioned ideal embodiment according to the present invention as inspiration, through the above description, relevant staff is complete
Various changes and amendments can be carried out without departing from the scope of the technological thought of the present invention' entirely.The technology of this invention
Property range is not limited to the contents of the specification, it is necessary to which the technical scope thereof is determined according to the scope of the claim.
Claims (6)
1. a kind of self study vibrating failure diagnosis method, which comprises the following steps:
(1) vibration signal is acquired in real time；
(2) by vibration signal by acquiring after AD conversion into self study vibrating failure diagnosis model；
(3) conversion of the time domain to frequency domain will be realized by Fourier transformation by the vibration signal after AD conversion, obtains vibration letter
Number vibration amplitude corresponding on a different frequency；
(4) it is state that any data are all failure by self study vibrating failure diagnosis model initialization, then learns by oneself determining
Practise vibrating failure diagnosis model it is normal in the state of, using by the frequency values in the vibration signal after Fourier transformation as neural
The first input value in network, it is defeated by first using vibration amplitude corresponding in frequency range as the second input value in neural network
Enter value, in the second input value input model, the output threshold range being calculated is replaced into self study vibrating failure diagnosis model
Middle initialization data is all the initial threshold range of failure, establishes diagnostic model；
(5) it keeps diagnostic model no longer to change, which as input signal and is inputted by the vibration signal that acquires in real time
In type, thus show whether Oscillation Amplitude is normal through judgement.
2. a kind of self study vibrating failure diagnosis method according to claim 1, it is characterised in that: in the step (1)
Vibration signal include Oscillation Amplitude signal or vibration audible signal in any one.
3. a kind of self study vibrating failure diagnosis method according to claim 1, it is characterised in that: in the step (4),
In the state of determining that self study vibrating failure diagnosis model is in normal by artificial or detecting instrument.
4. a kind of self study vibrating failure diagnosis method according to claim 1, it is characterised in that: in the step (4)
The second input value vibration amplitude corresponding in frequency range being re-used as after normalized in neural network.
5. a kind of self study vibrating failure diagnosis method according to claim 1, it is characterised in that: in the step (4)
First input value, the second input value are correspondingly by pairs of input model.
6. a kind of self study vibrating failure diagnosis method according to claim 1, it is characterised in that: in the step (5),
It as input signal and is inputted in the diagnostic model by the vibration signal that acquires in real time, when the result that diagnostic model is calculated
When in output threshold range, it is judged as that Oscillation Amplitude is normal；When the result that diagnostic model is calculated is beyond output threshold value
When range, it is judged as that Oscillation Amplitude is improper.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910116622.3A CN109738058A (en) | 2019-02-15 | 2019-02-15 | A kind of self study vibrating failure diagnosis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910116622.3A CN109738058A (en) | 2019-02-15 | 2019-02-15 | A kind of self study vibrating failure diagnosis method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109738058A true CN109738058A (en) | 2019-05-10 |
Family
ID=66367485
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910116622.3A Pending CN109738058A (en) | 2019-02-15 | 2019-02-15 | A kind of self study vibrating failure diagnosis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109738058A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113359210A (en) * | 2020-03-04 | 2021-09-07 | 罗伯特·博世有限公司 | Detecting when a piece of material is clamped between a chuck and a tool |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006113002A (en) * | 2004-10-18 | 2006-04-27 | Nsk Ltd | Anomaly diagnosis system for mechanical equipment |
| CN101858778A (en) * | 2010-05-28 | 2010-10-13 | 浙江大学 | Automatic fault diagnosis method of wind turbine based on vibration monitoring |
| CN103900816A (en) * | 2014-04-14 | 2014-07-02 | 上海电机学院 | Method for diagnosing bearing breakdown of wind generating set |
| CN105004498A (en) * | 2015-07-09 | 2015-10-28 | 西安理工大学 | Vibration fault diagnosis method of hydroelectric generating set |
| EP3322060A1 (en) * | 2016-11-14 | 2018-05-16 | Nordex Energy GmbH | Method of damping electromechanical oscillations on a power system |
| CN108241874A (en) * | 2018-02-13 | 2018-07-03 | 河南科技大学 | Video Text Region Location Method Based on BP Neural Network and Spectrum Analysis |
| CN108593286A (en) * | 2018-04-12 | 2018-09-28 | 中国神华能源股份有限公司 | The method for diagnosing faults of rotating machinery and the trouble-shooter of rotating machinery |
| CN108960077A (en) * | 2018-06-12 | 2018-12-07 | 南京航空航天大学 | A kind of intelligent failure diagnosis method based on Recognition with Recurrent Neural Network |
| CN109211548A (en) * | 2018-08-31 | 2019-01-15 | 沃德传动（天津）股份有限公司 | A kind of mechanical failure diagnostic method |
| CN109323754A (en) * | 2018-08-31 | 2019-02-12 | 南京理工大学 | A Method for Diagnosing and Detecting Polygon Faults of Train Wheels |
-
2019
- 2019-02-15 CN CN201910116622.3A patent/CN109738058A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006113002A (en) * | 2004-10-18 | 2006-04-27 | Nsk Ltd | Anomaly diagnosis system for mechanical equipment |
| CN101858778A (en) * | 2010-05-28 | 2010-10-13 | 浙江大学 | Automatic fault diagnosis method of wind turbine based on vibration monitoring |
| CN103900816A (en) * | 2014-04-14 | 2014-07-02 | 上海电机学院 | Method for diagnosing bearing breakdown of wind generating set |
| CN105004498A (en) * | 2015-07-09 | 2015-10-28 | 西安理工大学 | Vibration fault diagnosis method of hydroelectric generating set |
| EP3322060A1 (en) * | 2016-11-14 | 2018-05-16 | Nordex Energy GmbH | Method of damping electromechanical oscillations on a power system |
| CN108241874A (en) * | 2018-02-13 | 2018-07-03 | 河南科技大学 | Video Text Region Location Method Based on BP Neural Network and Spectrum Analysis |
| CN108593286A (en) * | 2018-04-12 | 2018-09-28 | 中国神华能源股份有限公司 | The method for diagnosing faults of rotating machinery and the trouble-shooter of rotating machinery |
| CN108960077A (en) * | 2018-06-12 | 2018-12-07 | 南京航空航天大学 | A kind of intelligent failure diagnosis method based on Recognition with Recurrent Neural Network |
| CN109211548A (en) * | 2018-08-31 | 2019-01-15 | 沃德传动（天津）股份有限公司 | A kind of mechanical failure diagnostic method |
| CN109323754A (en) * | 2018-08-31 | 2019-02-12 | 南京理工大学 | A Method for Diagnosing and Detecting Polygon Faults of Train Wheels |
Non-Patent Citations (12)
| Title |
|---|
| LUYANGJING: "《A convolutional neural network based feature learning and fault diagnosis method for the condition monitoring of gearbox》", 《MEASUREMENT》 * |
| YAN TAI-SHAN: "《An Improved Evolutionary Neural Network Algorithm and its Application in Fault Diagnosis for Hydropower Units》", 《2010 INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTATION TECHNOLOGY AND AUTOMATION》 * |
| YANHAN: "《Multi-level wavelet packet fusion in dynamic ensemble convolutional neural network for fault diagnosis》", 《MEASUREMENT》 * |
| 张明: "《基于动态自学习阈值和趋势滤波的机械故障智能预警方法》", 《振动与冲击》 * |
| 戚华峰: "《基于神经网络技术的核电厂阀门故障诊断》", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
| 李华等: "《设备维修与管理工程》", 31 March 2018 * |
| 王志永: "《基于振动信号频域分析法的铣齿机故障诊断》", 《制造技术与机床》 * |
| 王金福: "《机械故障诊断的信号处理方法： 频域分析》", 《噪声与振动控制》 * |
| 蔡艳平: "《基于小波包和神经网络的柴油机气门故障诊断》", 《机电工程技术》 * |
| 陈大禧: "《机械设备故障诊断基础知识》", 30 November 1989 * |
| 高晶波: "《基于神经网络的往复压缩机气阀故障诊断方法》", 《压缩机技术》 * |
| 魏中青: "《旋转机械振动监测快变报警系统研究及工程应用》", 《机电工程技术》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113359210A (en) * | 2020-03-04 | 2021-09-07 | 罗伯特·博世有限公司 | Detecting when a piece of material is clamped between a chuck and a tool |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6294268B2 (en) | Abnormality diagnosis device for injection molding machine | |
| SA522432736B1 (en) | Pipeline Sensor Integration for Product Mapping | |
| CN108731923A (en) | A kind of fault detection method and device of rotating machinery | |
| JP7462400B2 (en) | Artificial intelligence device and method for pre-processing noise data to identify problem noise sources | |
| US11830518B2 (en) | Sound data processing method, sound data processing device, and program | |
| JPWO2018061842A1 (en) | Abnormality detection program, abnormality detection method and abnormality detection device | |
| CN102662390B (en) | Fault diagnosis method of random fuzzy fault characteristic fusion rotating mechanical device | |
| CN108304348B (en) | A Bearing Remaining Life Prediction Method Based on Binary Wiener Process | |
| CN111504645B (en) | A fault diagnosis method of rolling bearing based on multi-point kurtosis in frequency domain | |
| JP2020191050A (en) | Abnormality detector and abnormality detection method | |
| EP3513066A1 (en) | Method and device for monitoring a status of at least one wind turbine and computer program product | |
| CN108447503A (en) | Motor Abnormal Noise Detection Method Based on Hilbert-Huang Transformation | |
| WO2019024450A1 (en) | Device fault detection method and apparatus | |
| CN109828168A (en) | Converter method for diagnosing faults based on Density Estimator | |
| CN109738058A (en) | A kind of self study vibrating failure diagnosis method | |
| CN106679847A (en) | Electric power equipment fault diagnosing method and apparatus | |
| CN110794329A (en) | Test method for defect identification ability of partial discharge live detector for combined electrical appliances and switchgear | |
| CN108398271A (en) | A kind of engine state monitor method based on vibration and nerual network technique | |
| CN111456915A (en) | Fault diagnosis device and method for internal components of fan engine room | |
| CN104390982A (en) | Test method for SMT first article inspection | |
| CN107576904B (en) | Fault Diagnosis Method of Analog Circuit Based on Sawtooth Wave | |
| CN103743477B (en) | Method and device for detecting and diagnosing mechanical faults | |
| TWI760904B (en) | Sound-based mechanical monitoring system and method | |
| JP7188463B2 (en) | ANALYSIS DEVICE, ANALYSIS METHOD, AND PROGRAM | |
| CN105740822B (en) | A kind of mechanical fault diagnosis optimization 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190510 |