CN106447040A - Method for evaluating the health state of mechanical equipment based on heterogeneous multi-sensor data fusion - Google Patents
Method for evaluating the health state of mechanical equipment based on heterogeneous multi-sensor data fusion Download PDFInfo
- Publication number
- CN106447040A CN106447040A CN201610865463.3A CN201610865463A CN106447040A CN 106447040 A CN106447040 A CN 106447040A CN 201610865463 A CN201610865463 A CN 201610865463A CN 106447040 A CN106447040 A CN 106447040A
- Authority
- CN
- China
- Prior art keywords
- signal
- sensor
- fault
- stress
- plant equipment
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a method for evaluating the health state of mechanical equipment based on heterogeneous multi-sensor data fusion. The method comprises the following steps: de-noising the signals of the heterogeneous multiple sensors; extracting the sensitive characteristic quantities of the respective signals; constructing a sensitive characteristic quantity set; using the sensitive characteristic quantity set as the training sample of a BP neural network; establishing a BP neural network fault diagnosis model to achieve fault separation and diagnosis; obtaining the recognition rates of respective faults of respective sensors; constructing a D-S evidence framework; and finally solving a maximum probability of failure type by using the Dempster synthesis formula of a D-S evidence theory. The method is based on the fusion of the D-S decision layer, makes the classification diagnosis high precision and high efficiency, and effectively improves a crack diagnosis effect, and is easy to be used in engineering practice.
Description
Technical field
The invention belongs to mechanical fault diagnosis technical field, more particularly to a kind of based on Heterogeneous Multi-Sensor Data
The plant equipment health state evaluation method of fusion.
Background technology
In plant equipment, incidence rate highest fault is crackle, and therefore, researcher is carried out to the Identification of Cracks of mechanical component
Substantial amounts of research, such as utilization based on acoustic emission signal in crack detection, using image processing techniquess and voice recognition
Both fusion methods of technology to the research of Ovum crusta Gallus domesticuss face crack automatic identification, using electromagnetic acoustic principle design a kind of pin
To steel plate flaw detection system；The utilization of the technology such as the crack detection based on leakage field technology.The above-mentioned technological means for referring to are right
All there are some constraints during the inspection of main equipment crackle, such as acoustic emission signal such as comparatively quietly will detect environment,
Machine vision is unable to real-time monitoring, and electromagnetic acoustic can not accomplish the factors such as microminiaturization.Accordingly, it is capable to the technology of real-time monitoring, still
It is vibrotechnique and strain gauge technology, accelerometer and strain gauge not only small volume, and high precision and can wireless and passive.Single
Sensor the detection of crackle had reliable believe the low shortcoming of low, precision.
Content of the invention
In order to solve above-mentioned technical problem, the present invention provide a kind of efficiency of the fault diagnosis to plant equipment with accurate
Property the high plant equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion.
The technical solution used in the present invention is：A kind of health state evaluation side of the data fusion based on Dissimilar sensors
Method, comprises the steps：
1) using different types of sensor, mechanical breakdown is measured, obtains vibration signal for faster and ess-strain letter
Number, to signal xiDenoising is carried out, obtains various kinds of sensors signal xi', wherein i represents signal from which sensor；
2) the sensitive features vector of time domain or/and frequency domain, for the measurement attribute of different types of sensor, is extracted respectively,
Constitute sensitive features vector set；
3) using the signal sensitive features amount of various kinds of sensors measurement as the input of BP neural network, BP nerve net is constructed
Network, adopts known sample to be trained to set up fault diagnosis model BP neural network；Recycle the multigroup of different faults
Experimental data is tested, and draws diagnosis of each sensor to each type fault；
4) D-S Evidence Framework is constructed according to the property value that the neutral net of each sensor draws, then adopts
Dempster composite formula, solves the probability fault type for drawing maximum.
In the above-mentioned plant equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion, step 1) tool
Body operating procedure is as follows：
1.1) select transducer arrangements point first, the sensor for measuring different physical attributes is respectively mounted corresponding biography
Sensor layout points, each sensor measurement obtains signal xi；
1.2) for acceleration signal, using singular value decomposition method denoising；For stress signal, first to its carrier wave
Signal is filtered, and then carries out singular value decomposition denoising again；Obtain the signal x after denoisingi'.
In the above-mentioned plant equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion, step 2) in adopt
Sensitive features vector set is constituted with the sensitive features vector of time domain, acceleration signal adopts p1,p2,p3As sensitive features vector；
Stress signal adopts p1,p2,p3,p4,p5,p6Used as sensitive features vector, strain signal adopts p1,p2,p3As sensitive features to
Amount；Wherein：p1=max (| xi|), In formula：For xiMeansigma methodss, xstdFor xiVariance.
In the above-mentioned plant equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion, step 3) in build
In vertical neural network model, including the property value G of the signal to mechanical equipment fault of each sensori=1 or 0,1 represents
Plant equipment is normal, and 0 represents mechanical equipment fault；And solve diagnosis of every class sensor for each mechanical breakdown
wi.
In the above-mentioned plant equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion, step 4) in D-
S Evidence Framework method for building up, using acceleration signal, stress signal and strain signal as three independent information sources, constitutes D-S
Evidence form；Judge that the normal credibility of plant equipment is P by acceleration signala1, plant equipment event is judged by acceleration signal
The credibility of barrier is Pa0；Judge that the normal credibility of plant equipment is P by stress signals1, plant equipment is judged by stress signal
The credibility of fault is Ps0；Judge that the normal credibility of plant equipment is P by strain signaln1, judge that machinery sets by strain signal
The credibility of standby fault is Pn0, and calculate acceleration signal and stress signal, strain signal for plant equipment normal,
Fault } class credibility Psan；The fault type of maximum likelihood is solved using Dempster composite formula.
Compared with prior art, the invention has the beneficial effects as follows：
The present invention carries out denoising respectively to the data of Dissimilar sensors first, then extracts the sensitivity of various signals
Characteristic quantity, is configured to vector set, using vector set as the training sample of BP neural network, sets up and is examined based on BP neural network fault
Disconnected model realization fault reconstruction and diagnosis, and discrimination of each sensor to the fault of each is drawn, it is configured to D-S evidence
Framework, finally using the Dempster composite formula of D-S evidence theory, solves the fault type for having drawn maximum likelihood.This
Invention is with following differences in the significant advantage of traditional method：
(1) measured using foreign peoples's multisensor and more object informations are obtained in that, can be examined from many aspects
Survey and diagnose；
(2) the same type of sensor fault model of BP neural network is set up, and constructs D-S Evidence Framework；
(3) fusion based on D-S decision-making level so that classification diagnosis high precision, efficiency high, effectively improves mechanical breakdown
Diagnostic result accuracy.
Description of the drawings
Fig. 1 is the flow chart of the present invention.
Fig. 2 is the state estimation procedure chart of the data fusion of the present invention.
Fig. 3 is that (it is crack position that A, B, C are transducer arrangements point, D point to cantilever beam sensor layout drawing, and crack length is
5mm, width is 0.5mm).
The stress time-domain oscillogram that Fig. 4 is measured for strain gauge.
Specific embodiment
The present invention is further illustrated below in conjunction with the accompanying drawings.
As shown in Figure 1, 2, the present invention includes following operating procedure：
1) using different types of sensor, mechanical breakdown is measured, acquisition vibration is directly gathered by sensor and is accelerated
Signal and stress, strain signal xi, then to signal xiDenoising is carried out, obtains various kinds of sensors signal xi', wherein i represents
Signal is from which sensor.During signals collecting, suitably passed according to the different choice of sensors measure physical attribute first
Sensor layout points, the sensor for measuring different physical attributes is separately mounted to corresponding transducer arrangements point.Signal denoising
During process, for acceleration signal, using singular value decomposition method denoising；For stress signal, first to its carrier signal
Filter, then carry out singular value decomposition denoising again；Obtain the signal x after denoisingi'.
2) for the measurement attribute of different types of sensor, time domain sensitive features vector sum frequency is extracted to each sensor
The sensitivity characteristic vector (can also only extract the sensitivity characteristic vector of time domain sensitive features vector or frequency domain) in domain, is constituted sensitive special
Levy vector set.Acceleration signal adopts p1,p2,p3As sensitive features vector；Stress signal adopts p1,p2,p3,p4,p5,p6Make
Vectorial for sensitive features, strain signal adopts p1,p2,p3As sensitive features vector；Wherein：p1=max (| xi|), Formula
In：For xiMeansigma methodss, xstdFor xiVariance.
Frequency domain sensitive features vector refers to frequency content, chooses 5 larger frequencies of amplitude, the letter that each sensor is measured
Number sensitive features vector extract, constitute sensitive features vector matrix, such as acceleration [wa1,wa2,...,wai], stress
[wss1,wss2,...,wssi], strain [wsn1,wsn2,...,wsni]；Then it is normalized respectively.
3) the sensitive features collection of different faults is carried out BP neural network study, sets up fault model；Recycle different events
100 groups of data of barrier are tested, and (i.e. priori is general for the diagnosis of every kind of fault to obtain 100 groups of data of different faults
Rate)；The type for probability is obtained by BP neural network per group data.
4) D-S evidence proposition is built with the signal of different types of sensor, is ordered as evidence using the prior probability that obtains
The content of topic, builds D-S Evidence Framework, builds D-S evidence table using diagnosis of each sensor for every kind of fault, by BP
Neutral net draws the property value G of the signal for mechanical equipment fault of each sensoriThe machinery that=1 or 0,1 is represented sets
Standby normal, 0 represents mechanical equipment fault.Using acceleration signal, stress signal and stress and acceleration signal as three independences
Information source, therefore may be constructed D-S evidence form.
For example：When accelerometer property value is that 1, stress property value is 0, its evidence table is：
Judge that the normal credibility of plant equipment is P by acceleration signala1, mechanical equipment fault is judged by acceleration signal
Credibility be Pa0；Judge that the normal credibility of plant equipment is P by stress signals1, plant equipment event is judged by stress signal
The credibility of barrier is Ps0；Judge that the normal credibility of plant equipment is P by stress and acceleration signaln1, by stress and acceleration
Signal judges that the credibility of mechanical equipment fault is Pn0, and calculate acceleration signal and stress signal, stress and acceleration signal
{ normal, the fault } class for plant equipment credibility Psan；The composition rule is：
In formula：m1(A1)、m2(A2)、m3(A3) belief function, A1、A2、A3Refer to fault type.The probability that synthesis is obtained enters
Row compares, when normally differing larger with the probability of fault, then it is assumed that probability more greatly testing result；Probability when normal and fault
It is more or less the same, both ratio is in 0.8-1.2, then it is assumed that this judged result cannot be given, at this moment thinks that result is undetermined, needs
Manually to be detected, accurate result is given, and re-starts the training of BP nerve net.
Application example
With the cantilever beam shown in Fig. 3, Fig. 4 as object, select containing 25 groups of crackle cantilever beam signal and normal cantilever beam
25 groups of signal, and according to the sensitive features vector for being extracted herein above, form sensitive features vector set, totally 50 groups of conducts
Training sample.30 groups of reselection contains crackle cantilever beam signal and 20 groups of normal cantilever beam signals as test sample, is tied
Fruit is as follows：
Table 1
From table 1 it is recognised that based on accelerometer or the accuracy of the similar multi-Sensor Information Fusion Approach of strain gauge
For 74% and 86%, and the accuracy based on accelerometer and strain gauge neural Network Data Fusion method is 90% and based on god
The correct of D-S evidence theory Decision fusion method through network characterization identification is 96%.So Heterogeneous Multi-Sensor Data fusion
There is higher accuracy.
Claims (5)
1. a kind of plant equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion, comprises the steps：
1) using different types of sensor, mechanical breakdown is measured, vibration signal for faster and ess-strain signal is obtained,
To signal xiDenoising is carried out, obtains various kinds of sensors signal xi', wherein i represents signal from which sensor；
2) for the measurement attribute of different types of sensor, extract the sensitive features vector of time domain or/and frequency domain respectively, constitute
Sensitive features vector set；
3) using the signal sensitive features amount of various kinds of sensors measurement as the input of BP neural network, BP neural network is constructed, is adopted
With known sample, BP neural network is trained to set up fault diagnosis model；Recycle multigroup experiment number of different faults
According to being tested, diagnosis of each sensor to each type fault are drawn；
4) D-S Evidence Framework is constructed according to the property value that the neutral net of each sensor draws, then is closed using Dempster
Become formula, solve the probability fault type for drawing maximum.
2. the plant equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion according to claim 1,
Step 1) concrete operation step as follows：
1.1) select transducer arrangements point first, the sensor for measuring different physical attributes is respectively mounted corresponding sensor
Layout points, each sensor measurement obtains signal xi；
1.2) for acceleration signal, using singular value decomposition method denoising；For stress signal, first to its carrier signal
Filter, then carry out singular value decomposition denoising again；Obtain the signal x after denoisingi'.
3. the plant equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion according to claim 1,
Step 2) in sensitive features vector set is constituted using the sensitive features vector of time domain, acceleration signal adopts p1,p2,p3As quick
Sense characteristic vector；Stress signal adopts p1,p2,p3,p4,p5,p6Used as sensitive features vector, strain signal adopts p1,p2,p3Make
For sensitive features vector；Wherein：p1=max (| xi|), In formula：For xiMeansigma methodss, xstdFor xiVariance.
4. the plant equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion according to claim 1,
Step 3) in the neural network model set up, including the property value G of the signal to mechanical equipment fault of each sensori=
1or 0,1 represents that plant equipment is normal, and 0 represents mechanical equipment fault；And every class sensor is solved for each machinery event
The diagnosis w of barrieri.
5. the plant equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion according to claim 4,
Step 4) in D-S Evidence Framework method for building up, using acceleration signal, stress signal, stress and acceleration signal as three solely
Vertical information source, constitutes D-S evidence form；Judge that the normal credibility of plant equipment is P by acceleration signala1, by acceleration
Signal judges that the credibility of mechanical equipment fault is Pa0；Judge that the normal credibility of plant equipment is P by stress signals1, by should
Force signal judges that the credibility of mechanical equipment fault is Ps0；The normal credibility of plant equipment is judged by stress and acceleration signal
For Pn1, the credibility for judging mechanical equipment fault by stress and acceleration signal is Pn0, and calculate acceleration signal and stress,
Credibility P of { normal, fault } class for plant equipment of stress and acceleration signalsan；Using Dempster composite formula
Solve the fault type of maximum likelihood.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610865463.3A CN106447040B (en) | 2016-09-30 | 2016-09-30 | Mechanical equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610865463.3A CN106447040B (en) | 2016-09-30 | 2016-09-30 | Mechanical equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106447040A true CN106447040A (en) | 2017-02-22 |
| CN106447040B CN106447040B (en) | 2018-11-23 |
Family
ID=58170042
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610865463.3A Active CN106447040B (en) | 2016-09-30 | 2016-09-30 | Mechanical equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106447040B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108932581A (en) * | 2018-06-06 | 2018-12-04 | 深圳大学 | The autonomous cognitive method and system of more physics domain information fusions |
| CN109254526A (en) * | 2018-09-06 | 2019-01-22 | 南京航空航天大学 | A kind of multilevel security redundancy control system hanging voluntarily transport trolley |
| CN109421743A (en) * | 2017-08-28 | 2019-03-05 | 通用汽车环球科技运作有限责任公司 | Method and apparatus for monitoring autonomous vehicle |
| WO2019080367A1 (en) * | 2017-10-25 | 2019-05-02 | 南京航空航天大学 | Method for evaluating health status of mechanical device |
| CN110188378A (en) * | 2019-04-12 | 2019-08-30 | 浙江大学 | A kind of aerodynamic data fusion method neural network based |
| CN110553770A (en) * | 2019-07-17 | 2019-12-10 | 哈尔滨工程大学 | Fault diagnosis algorithm for large drift abnormal value of sensor of ship structure stress monitoring system |
| CN112598019A (en) * | 2020-10-28 | 2021-04-02 | 南京航空航天大学 | Heterogeneous vibration signal space-time multi-cross correlation analysis and self-adaptive weighting fusion method |
| CN112857669A (en) * | 2021-03-30 | 2021-05-28 | 武汉飞恩微电子有限公司 | Fault detection method, device and equipment of pressure sensor and storage medium |
| CN114690038A (en) * | 2022-06-01 | 2022-07-01 | 华中科技大学 | Motor fault identification method and system based on neural network and storage medium |
| CN115290317A (en) * | 2022-06-24 | 2022-11-04 | 西南交通大学 | Industrial robot speed reducer health state assessment method based on multi-source information fusion |
| CN115687984A (en) * | 2023-01-05 | 2023-02-03 | 绍兴市特种设备检测院 | Method for monitoring health state of stirring kettle |
| CN116881846A (en) * | 2023-07-17 | 2023-10-13 | 无锡北微传感科技有限公司 | Multi-mode communication iron tower monitoring method based on multi-sensor information fusion |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101088651B1 (en) * | 2009-09-30 | 2011-12-01 | 성균관대학교산학협력단 | Method and apparatus for context estimating |
| CN102589890A (en) * | 2012-03-01 | 2012-07-18 | 上海电力学院 | Integrated fault diagnostic method of steam turbine based on CPN (counter-propagation network) and D-S (dempster-shafer) evidences |
| CN102928231A (en) * | 2012-11-13 | 2013-02-13 | 上海电力学院 | Equipment fault diagnosis method based on D-S (Dempster-Shafer) evidence theory |
| CN105675274A (en) * | 2016-01-07 | 2016-06-15 | 西安交通大学 | Time-domain parameter and D-S evidence theory-based rotor running state monitoring method |
-
2016
- 2016-09-30 CN CN201610865463.3A patent/CN106447040B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101088651B1 (en) * | 2009-09-30 | 2011-12-01 | 성균관대학교산학협력단 | Method and apparatus for context estimating |
| CN102589890A (en) * | 2012-03-01 | 2012-07-18 | 上海电力学院 | Integrated fault diagnostic method of steam turbine based on CPN (counter-propagation network) and D-S (dempster-shafer) evidences |
| CN102928231A (en) * | 2012-11-13 | 2013-02-13 | 上海电力学院 | Equipment fault diagnosis method based on D-S (Dempster-Shafer) evidence theory |
| CN105675274A (en) * | 2016-01-07 | 2016-06-15 | 西安交通大学 | Time-domain parameter and D-S evidence theory-based rotor running state monitoring method |
Non-Patent Citations (3)
| Title |
|---|
| 万林等: "基于人工神经网络和证据理论的电梯健康状态评估", 《工业控制计算机》 * |
| 曾文韬等: "基于BP神经网络混凝土泵车臂架故障预测研究", 《机械设计与制造》 * |
| 滑广军: "混凝土泵车臂架结构健康监测关键技术研究", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109421743A (en) * | 2017-08-28 | 2019-03-05 | 通用汽车环球科技运作有限责任公司 | Method and apparatus for monitoring autonomous vehicle |
| WO2019080367A1 (en) * | 2017-10-25 | 2019-05-02 | 南京航空航天大学 | Method for evaluating health status of mechanical device |
| CN108932581A (en) * | 2018-06-06 | 2018-12-04 | 深圳大学 | The autonomous cognitive method and system of more physics domain information fusions |
| CN109254526A (en) * | 2018-09-06 | 2019-01-22 | 南京航空航天大学 | A kind of multilevel security redundancy control system hanging voluntarily transport trolley |
| CN110188378A (en) * | 2019-04-12 | 2019-08-30 | 浙江大学 | A kind of aerodynamic data fusion method neural network based |
| CN110553770A (en) * | 2019-07-17 | 2019-12-10 | 哈尔滨工程大学 | Fault diagnosis algorithm for large drift abnormal value of sensor of ship structure stress monitoring system |
| CN110553770B (en) * | 2019-07-17 | 2021-03-26 | 哈尔滨工程大学 | Fault diagnosis method for large drift abnormal value of sensor of ship structure stress monitoring system |
| CN112598019B (en) * | 2020-10-28 | 2024-04-02 | 南京航空航天大学 | Heterogeneous vibration signal space-time multi-cross-correlation analysis and self-adaptive weighted fusion method |
| CN112598019A (en) * | 2020-10-28 | 2021-04-02 | 南京航空航天大学 | Heterogeneous vibration signal space-time multi-cross correlation analysis and self-adaptive weighting fusion method |
| CN112857669A (en) * | 2021-03-30 | 2021-05-28 | 武汉飞恩微电子有限公司 | Fault detection method, device and equipment of pressure sensor and storage medium |
| CN112857669B (en) * | 2021-03-30 | 2022-12-06 | 武汉飞恩微电子有限公司 | Fault detection method, device and equipment of pressure sensor and storage medium |
| CN114690038A (en) * | 2022-06-01 | 2022-07-01 | 华中科技大学 | Motor fault identification method and system based on neural network and storage medium |
| CN115290317A (en) * | 2022-06-24 | 2022-11-04 | 西南交通大学 | Industrial robot speed reducer health state assessment method based on multi-source information fusion |
| CN115687984A (en) * | 2023-01-05 | 2023-02-03 | 绍兴市特种设备检测院 | Method for monitoring health state of stirring kettle |
| CN116881846A (en) * | 2023-07-17 | 2023-10-13 | 无锡北微传感科技有限公司 | Multi-mode communication iron tower monitoring method based on multi-sensor information fusion |
| CN116881846B (en) * | 2023-07-17 | 2024-04-05 | 无锡北微传感科技有限公司 | Multi-mode communication iron tower monitoring method based on multi-sensor information fusion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106447040B (en) | 2018-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106447040B (en) | Mechanical equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion | |
| CN104665849B (en) | A kind of high ferro dispatcher based on the interaction of many physiological signals multi-model stress detection method | |
| CN102419437B (en) | Track-before-detect method based on flight path inspection | |
| CN104316729B (en) | Self-diagnosis method of acceleration sensors for locomotive bogie detection | |
| CN107014668A (en) | A kind of fatigue crack integrated monitoring based on piezoelectricity and smart coat sensor | |
| CN108562821B (en) | Method and system for determining single-phase earth fault line selection of power distribution network based on Softmax | |
| CN106247173B (en) | The method and device of pipeline leakage testing | |
| CN112101174A (en) | LOF-Kurtogram-based mechanical fault diagnosis method | |
| CN111504645B (en) | Rolling bearing fault diagnosis method based on frequency domain multipoint kurtosis | |
| CN111678699B (en) | Early fault monitoring and diagnosing method and system for rolling bearing | |
| CN109406147A (en) | Train bearing rail side acoustic diagnosis method under variable speed working condition | |
| CN109375010A (en) | Winding in Power Transformer method for diagnosing faults based on vibration distribution characteristic model | |
| CN106228107B (en) | A kind of supersonic guide-wave broken rail monitoring method based on independent component analysis | |
| CN109932179A (en) | A kind of rolling bearing fault testing method based on the reconstruct of DS Adaptive spectra | |
| CN105675274A (en) | Time-domain parameter and D-S evidence theory-based rotor running state monitoring method | |
| CN106338662A (en) | Transformer winding deformation diagnosis method based on mathematical morphology | |
| WO2021248962A1 (en) | Non-destructive testing method and device for testing and distinguishing internal and external defects of steel wire rope | |
| CN106708009A (en) | Ship dynamic positioning measurement system multiple-fault diagnosis method based on support vector machine clustering | |
| CN103267652B (en) | Intelligent online diagnosis method for early failures of equipment | |
| CN110553789A (en) | state detection method and device of piezoresistive pressure sensor and brake system | |
| CN106569160A (en) | Output voltage sensor fault diagnosing method of auxiliary current transformer and fault tolerance control method | |
| CN108896308A (en) | A kind of wheel set bearing method for diagnosing faults based on probability envelope | |
| CN109827081A (en) | A kind of buried drain pipe road plugging fault and branch pipe tee connection part diagnostic method based on acoustics active detecting | |
| CN114462480A (en) | Multi-source sensor rolling mill fault diagnosis method based on non-equilibrium data set | |
| CN107355687A (en) | A kind of sewer pipe fault detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |