CN105157821A - Rotary machinery vibration fault diagnosis and quantitative analysis method - Google Patents
Rotary machinery vibration fault diagnosis and quantitative analysis method Download PDFInfo
- Publication number
- CN105157821A CN105157821A CN201510539779.9A CN201510539779A CN105157821A CN 105157821 A CN105157821 A CN 105157821A CN 201510539779 A CN201510539779 A CN 201510539779A CN 105157821 A CN105157821 A CN 105157821A
- Authority
- CN
- China
- Prior art keywords
- fault
- signal
- diagnosis
- sample
- information entropy
- 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
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to a rotary machinery vibration fault diagnosis and quantitative analysis method. When a series of signal analysis methods and model establishing platforms are formed, examination standards for models are not introduced into the models; and there is no quantitative analysis standard for severity of faults. The method comprises the following steps that accurate diagnosis of the faults is realized with diagnosis time consumption and diagnosis accuracy acting as measurement standards, an ensemble empirical mode decomposition method is selected to perform signal decomposition in view of nonlinearity of vibration signals and large amount of abnormal signals included in the signals, a model is established on a support vector machine platform with intrinsic mode energy acting as characteristics and diagnosis time consumption and accuracy acting as the measurement standards so that the model is enabled to be better suitable for rapid and effective diagnosis of the vibration faults. A complete set of methods from signal analysis, characteristics extraction, fault diagnosis to fault quantitative measurement are put forward, and accuracy is high.
Description
Technical field
The present invention relates to a kind of rotating machinery vibrating failure diagnosis and quantitative analysis method, specifically realize the method for vibration fault intelligent diagnostics and the quantitative test of the vibration fault order of severity.
Background technology
While plant equipment development, people have higher requirement to the reliability of equipment and security, and device security reliability service becomes the important topic of current equipment operation and maintenance.Effective facilities and equipments state-detection and fault diagnosis, by initial stage sign identification fault and then can take corresponding solution, avoid the generation of great serious accident, support equipment safe operation.
Also have some about the real-time diagnosis method of equipment now, if publication date is on 05 13rd, 2009, publication number is in the Chinese patent of CN101430247, discloses a kind of real-time diagnosis method for random vibration fault of steam generator set; And publication date is on 09 07th, 2011, publication number is in the Chinese patent of CN102175307A, discloses a kind of real-time quantitative analysis method for low-frequency vibration spectrum array of steam turbine generator set.These methods are difficult to realize implementing fault diagnosis by initial stage sign, and then realize the quantitative measurement analysis of fault severity level.
How realizing fault diagnosis is fast and effectively current main research direction, current a series of signal analysis method and model buildings platform molded, but to introduce not yet in a model for the standard of considering of model; Simultaneously for fault severity level, not yet have quantitative test standard, often have the operation pressure and shut down the contradiction between processing and exist, therefore whether fault processes in time, is the problem needing to determine after fault diagnosis.
Summary of the invention
The object of the invention is to overcome above shortcomings in prior art, and a kind of rotating machinery vibrating failure diagnosis and quantitative analysis method are provided.Propose from signal analysis, feature extraction, fault diagnosis to the overall procedure of fault quantitative measurement.
The present invention's adopted technical scheme that solves the problem is: the feature of this rotating machinery vibrating failure diagnosis and quantitative analysis method is: the step of described method is as follows: to diagnose consuming time and accuracy rate of diagnosis for criterion is to realize fault Precise Diagnosis, a large amount of abnormal signal is included for the non-linear of vibration signal and signal, choose set empirical mode decomposition method and carry out signal decomposition, support vector machine platform establishes one with intrinsic mode energy for feature, to diagnose the model being criterion with accuracy rate consuming time, it is made well to adapt to vibration fault quick, efficient diagnosis.
As preferably, the present invention is directed to on-the-spot vibration signal and first carried out signal denoising analysis and research, determining with signal to noise ratio (S/N ratio) is the denoising scheme of criterion.
As preferably, the present invention is by carrying out signal gathering the Intrinsic Mode energy feature that empirical mode decomposition extracts vibration signal.
As preferably, the present invention, on support vector machine platform, with the fault signature extracted for sample, carries out diagnosis validation.
As preferably, the present invention is on the basis of vibration fault Accurate Diagnosis, and the concept introducing information entropy carries out the measurement of the vibration fault order of severity, for dissimilar vibration fault comformed information entropy parameter quantification range.
As preferably, the present invention carries out signal sampling setting according to signal analysis law, collection site unit operation vibration signal.
Introduce signal to noise ratio (S/N ratio) concept and carry out based Denoising to vibration signal, by carrying out grid optimizing to EEMD resolution parameter, determine the optimization model of vibration signal denoising, optimal Decomposition parameter and standard differs from 0.6, adds noise number of times 200.
Carry out fault signature extraction for the signal after denoising, determine that fault signature extracts most effective scheme.Set empirical mode decomposition being carried out to signal, carrying out power feature extraction to decomposing the Intrinsic mode function obtained, composition fault feature vector.
The proper vector be directly made up of Intrinsic Mode energy has more existence of redundant, carries out dimension-reduction treatment by PCA and RS, to realize quick efficient diagnosis.
Support vector machine builds fault diagnosis model, as training and testing object, model is verified using the fault signature extracted.
With the fault signature before and after dimensionality reduction for sample, carry out fault diagnosis model checking.Proof procedure is weighed to diagnose the consuming time and double standard of accuracy rate of diagnosis, and the distinct methods determined the parameter in model process of establishing respectively carries out analysiss and contrasts.
Determine the vibrating failure diagnosis model that the Intrinsic Mode energy after with dimensionality reduction is feature, all there is clear superiority in it on the consuming time and accuracy rate of diagnosis of diagnosis.
Introduce the concept of information entropy, determine the order of severity of vibration fault used as quantitative measurement.
Single vibration fault sample before dimensionality reduction is arranged, carries out the calculating of information entropy parameter for the fault signature be made up of IMF energy.
By carrying out association analysis to the fault signature component in information entropy result and fault sample, sum up the rule that fault severity level follows information entropy Parameters variation, and carry out statistical study for the information entropy parameter area of different faults.
As preferably, the step of the method for the invention is as follows:
Step 1: according to mass unbalance, impact and rub, misalign, oil whirl selects 20 groups of fault samples;
Step 2: carry out based Denoising optimization of parameter choice to signal, finally determines EEMD signal denoising scheme;
Step 3: carry out denoising normative reference with mutual information and zero-crossing rate dual indexes, determine that mutual information threshold values is 0.4, zero-crossing rate threshold values is 0.6, at utmost ensures original signal with the signal after this threshold values process;
Step 4: respectively EEMD decomposition is carried out to 20 groups of signals, and analyze each fault each sample IMF component Failure Distribution;
Step 5: each sample IMF component is made energy calculation, and forms fault sample energy feature with this;
Step 6: carry out redundant information process to fault sample feature, is analyzed by PCA and RS, constitutes the fault sample feature after dimension-reduction treatment;
Step 7: set up fault diagnosis model on support vector machine platform, model Kernel Function two parameter determination schemes select grid optimization, GA to optimize simultaneously and PSO optimization is compared, and carry out modelling verification with the fault sample extracted above;
Step 8: to diagnose consuming time and accuracy rate of diagnosis for criterion, carry out the criterion of fault diagnosis model, first analyzing and processing is being carried out to the fault sample before and after dimensionality reduction in step 6, determine sample PCA dimension-reduction treatment method by parameter;
Step 9: with the sample before and after dimensionality reduction for contrast object, carry out checking respectively select kind of the optimization of parameter choice method of three in step 7, with these 20 groups of sample checkings, draws the IMF energy feature after dimensionality reduction;
Step 10: introduce information entropy parameter, carry out fault sample analysis, define according to information entropy the distribution that it can characterize quantity of information in fault sample very well, chooses imbalance fault sample and has carried out the calculating of preliminary information entropy;
Step 11: respectively information entropy calculating is carried out to four kinds of faults, draw: for misaligning, impact and rub and oil whirl three kinds of faults are along with the increase of fault severity level, information entropy parameter increases gradually, when information entropy parameter is greater than 1.5, represents that fault degree is more serious; For quality imbalance fault, its fault degree is more serious, and information entropy parameter is less, when information entropy parameter is less than 0.3, characterizes fault more serious.
The present invention compared with prior art, has the following advantages and effect: mainly for the steam turbine operated under current high parameter, high rotating speed, implements fault diagnosis, and then realize the quantitative measurement analysis of fault severity level by initial stage sign.The present invention defines the integral framework of signal analysis, denoising, feature extraction, diagnostic model optimization, fault severity level quantitative measurement.Come from the noise that vibration signal mixes, determine signal denoising scheme by signal analysis; Determine EEMD decomposition method and carry out decomposition extraction feature for signal, establish to diagnose the Precise Diagnosis being criterion determination optimal diagnosis model realization vibration fault with accuracy rate of diagnosis consuming time, for the quantitative measurement vibration fault order of severity introduces the concept of information entropy, for different vibration fault comformed information entropy parameter scope.Design of the present invention is unique, and simple to operate, accuracy rate is high.
Accompanying drawing explanation
Optimization of parameter choice process when Fig. 1 is a certain vibration signal based Denoising in the embodiment of the present invention.
Fig. 2 is that in the embodiment of the present invention, a certain vibration signal carries out each IMF component after EEMD decomposition.
Embodiment
Below in conjunction with accompanying drawing, also by embodiment, the present invention is described in further detail, and following examples are explanation of the invention and the present invention is not limited to following examples.
Embodiment.
See Fig. 1 to Fig. 2, in the present embodiment, the step of rotating machinery vibrating failure diagnosis and quantitative analysis method is as follows: to diagnose consuming time and accuracy rate of diagnosis for criterion is to realize fault Precise Diagnosis, a large amount of abnormal signal is included for the non-linear of vibration signal and signal, choose set empirical mode decomposition method and carry out signal decomposition, support vector machine platform establishes one with intrinsic mode energy for feature, to diagnose the model being criterion with accuracy rate consuming time, vibration fault is quick, efficient diagnosis to make it well adapt to.
First carried out signal denoising analysis and research for on-the-spot vibration signal in the present embodiment, determining with signal to noise ratio (S/N ratio) is the denoising scheme of criterion.By carrying out signal gathering the Intrinsic Mode energy feature that empirical mode decomposition extracts vibration signal.On support vector machine platform, with the fault signature extracted for sample, carry out diagnosis validation.On the basis of vibration fault Accurate Diagnosis, the concept introducing information entropy carries out the measurement of the vibration fault order of severity, for dissimilar vibration fault comformed information entropy parameter quantification range.
Signal sampling setting is carried out according to signal analysis law, collection site unit operation vibration signal in the present embodiment; Introduce signal to noise ratio (S/N ratio) concept and based Denoising is carried out to vibration signal, by carrying out grid optimizing to EEMD resolution parameter, determining the model of vibration signal denoising, obtaining resolution parameter standard deviation 0.6, add noise number of times 200; Carrying out fault signature extraction for the signal after denoising, determine fault signature extraction scheme, set empirical mode decomposition is carried out to signal, carrying out power feature extraction to decomposing the Intrinsic mode function obtained, composition fault feature vector; The proper vector be directly made up of Intrinsic Mode energy has more existence of redundant, carries out dimension-reduction treatment by PCA and RS, to realize quick efficient diagnosis; Support vector machine builds fault diagnosis model, as training and testing object, model is verified using the fault signature extracted; With the fault signature before and after dimensionality reduction for sample, carry out fault diagnosis model checking.Proof procedure is weighed to diagnose the consuming time and double standard of accuracy rate of diagnosis, and the distinct methods determined the parameter in model process of establishing respectively carries out analysiss and contrasts; Determine the vibrating failure diagnosis model that the Intrinsic Mode energy after with dimensionality reduction is feature; Introduce the concept of information entropy, determine the order of severity of vibration fault used as quantitative measurement; Single vibration fault sample before dimensionality reduction is arranged, carries out the calculating of information entropy parameter for the fault signature be made up of IMF energy; By carrying out association analysis to the fault signature component in information entropy result and fault sample, sum up the rule that fault severity level follows information entropy Parameters variation, and carry out statistical study for the information entropy parameter area of different faults.
In the present embodiment, the step of the concrete grammar of rotating machinery vibrating failure diagnosis and quantitative analysis method is as follows.
Step 1: according to mass unbalance, impact and rub, misalign, oil whirl selects 20 groups of fault samples.
Step 2: carry out based Denoising optimization of parameter choice to signal according to the method in Fig. 2, finally determines EEMD signal denoising scheme.
Step 3: carry out denoising normative reference with mutual information and zero-crossing rate dual indexes, determine that mutual information threshold values is 0.4, zero-crossing rate threshold values is 0.6.At utmost original signal is ensured with the signal after this threshold values process.
Step 4: respectively EEMD decomposition is carried out to 20 groups of signals, and analyze each fault each sample IMF component Failure Distribution.
Step 5: each sample IMF component is made energy calculation, and forms fault sample energy feature with this.
Step 6: carry out redundant information process to fault sample feature, is analyzed by PCA and RS, constitutes the fault sample feature after dimension-reduction treatment.
Step 7: set up fault diagnosis model on support vector machine platform.Model Kernel Function two parameter determination schemes select grid optimization, GA optimization simultaneously, PSO optimizes and compares, and carry out modelling verification with the fault sample extracted above.
Step 8: diagnosing consuming time is criterion with accuracy rate of diagnosis, carries out the criterion of fault diagnosis model, is first carrying out analyzing and processing to the fault sample before and after dimensionality reduction in step 6, determine sample PCA dimension-reduction treatment method by parameter.
Step 9: with the sample before and after dimensionality reduction for contrast object, carries out checking to 3 kinds of optimization of parameter choice methods in step 7 respectively and selects, as shown in table 1 with the result of these 20 groups of sample checkings.As can be seen from Table 1, the IMF energy feature after dimensionality reduction, can obtain higher accuracy rate of diagnosis under support vector machine parametric grid cross validation scheme, and consuming time less.
Table 1
Step 10: introduce information entropy parameter, carry out fault sample analysis, define according to information entropy the distribution that it can characterize quantity of information in fault sample very well, choose imbalance fault sample and carried out the calculating of preliminary information entropy, result of calculation is as shown in table 2.
Table 2
Step 11: respectively information entropy calculating has been carried out to four kinds of faults, and be summarized as follows rule: for misaligning, impact and rub, oil whirl three kinds of faults are along with the increase of fault severity level, information entropy parameter increases gradually, when information entropy parameter is greater than 1.5, represent that fault degree is more serious; For quality imbalance fault, its fault degree is more serious, and information entropy parameter is less, when information entropy parameter is less than 0.3, characterizes fault more serious.
In addition, it should be noted that, the specific embodiment described in this instructions, the shape, institute's title of being named etc. of its parts and components can be different, and the above content described in this instructions is only to structure example of the present invention explanation.The equivalence change that structure, feature and the principle of all foundations described in inventional idea of the present invention are done or simple change, be included in the protection domain of patent of the present invention.Those skilled in the art can make various amendment or supplement or adopt similar mode to substitute to described specific embodiment; only otherwise depart from structure of the present invention or surmount this scope as defined in the claims, protection scope of the present invention all should be belonged to.
Claims (7)
1. a rotating machinery vibrating failure diagnosis and quantitative analysis method, it is characterized in that: the step of described method is as follows: to diagnose consuming time and accuracy rate of diagnosis for criterion is to realize fault Precise Diagnosis, a large amount of abnormal signal is included for the non-linear of vibration signal and signal, choose set empirical mode decomposition method and carry out signal decomposition, support vector machine platform establishes one with intrinsic mode energy for feature, to diagnose the model being criterion with accuracy rate consuming time, vibration fault is quick, efficient diagnosis to make it well adapt to.
2. rotating machinery vibrating failure diagnosis according to claim 1 and quantitative analysis method, is characterized in that: first carried out signal denoising analysis and research for on-the-spot vibration signal, and determining with signal to noise ratio (S/N ratio) is the denoising scheme of criterion.
3. rotating machinery vibrating failure diagnosis according to claim 1 and quantitative analysis method, is characterized in that: by carrying out signal gathering the Intrinsic Mode energy feature that empirical mode decomposition extracts vibration signal.
4. rotating machinery vibrating failure diagnosis according to claim 1 and quantitative analysis method, is characterized in that: on support vector machine platform, with the fault signature extracted for sample, carries out diagnosis validation.
5. rotating machinery vibrating failure diagnosis according to claim 1 and quantitative analysis method, it is characterized in that: on the basis of vibration fault Accurate Diagnosis, the concept introducing information entropy carries out the measurement of the vibration fault order of severity, for dissimilar vibration fault comformed information entropy parameter quantification range.
6. rotating machinery vibrating failure diagnosis according to claim 1 and quantitative analysis method, is characterized in that: carry out signal sampling setting according to signal analysis law, collection site unit operation vibration signal; Introduce signal to noise ratio (S/N ratio) concept and based Denoising is carried out to vibration signal, by carrying out grid optimizing to EEMD resolution parameter, determining the model of vibration signal denoising, obtaining resolution parameter standard deviation 0.6, add noise number of times 200; Carrying out fault signature extraction for the signal after denoising, determine fault signature extraction scheme, set empirical mode decomposition is carried out to signal, carrying out power feature extraction to decomposing the Intrinsic mode function obtained, composition fault feature vector; The proper vector be directly made up of Intrinsic Mode energy has more existence of redundant, carries out dimension-reduction treatment by PCA and RS, to realize quick efficient diagnosis; Support vector machine builds fault diagnosis model, as training and testing object, model is verified using the fault signature extracted; With the fault signature before and after dimensionality reduction for sample, carry out fault diagnosis model checking; Proof procedure is weighed to diagnose the consuming time and double standard of accuracy rate of diagnosis, and the distinct methods determined the parameter in model process of establishing respectively carries out analysiss and contrasts; Determine the vibrating failure diagnosis model that the Intrinsic Mode energy after with dimensionality reduction is feature; Introduce the concept of information entropy, determine the order of severity of vibration fault used as quantitative measurement; Single vibration fault sample before dimensionality reduction is arranged, carries out the calculating of information entropy parameter for the fault signature be made up of IMF energy; By carrying out association analysis to the fault signature component in information entropy result and fault sample, sum up the rule that fault severity level follows information entropy Parameters variation, and carry out statistical study for the information entropy parameter area of different faults.
7. rotating machinery vibrating failure diagnosis according to claim 1 and quantitative analysis method, is characterized in that: the step of described method is as follows:
Step 1: according to mass unbalance, impact and rub, misalign, oil whirl selects 20 groups of fault samples;
Step 2: carry out based Denoising optimization of parameter choice to signal, finally determines EEMD signal denoising scheme;
Step 3: carry out denoising normative reference with mutual information and zero-crossing rate dual indexes, determine that mutual information threshold values is 0.4, zero-crossing rate threshold values is 0.6, at utmost ensures original signal with the signal after this threshold values process;
Step 4: respectively EEMD decomposition is carried out to 20 groups of signals, and analyze each fault each sample IMF component Failure Distribution;
Step 5: each sample IMF component is made energy calculation, and forms fault sample energy feature with this;
Step 6: carry out redundant information process to fault sample feature, is analyzed by PCA and RS, constitutes the fault sample feature after dimension-reduction treatment;
Step 7: set up fault diagnosis model on support vector machine platform, model Kernel Function two parameter determination schemes select grid optimization, GA to optimize simultaneously and PSO optimization is compared, and carry out modelling verification with the fault sample extracted above;
Step 8: to diagnose consuming time and accuracy rate of diagnosis for criterion, carry out the criterion of fault diagnosis model, first analyzing and processing is being carried out to the fault sample before and after dimensionality reduction in step 6, determine sample PCA dimension-reduction treatment method by parameter;
Step 9: with the sample before and after dimensionality reduction for contrast object, carry out checking respectively select kind of the optimization of parameter choice method of three in step 7, with these 20 groups of sample checkings, draws the IMF energy feature after dimensionality reduction;
Step 10: introduce information entropy parameter, carry out fault sample analysis, define according to information entropy the distribution that it can characterize quantity of information in fault sample very well, chooses imbalance fault sample and has carried out the calculating of preliminary information entropy;
Step 11: respectively information entropy calculating is carried out to four kinds of faults, draw: for misaligning, impact and rub and oil whirl three kinds of faults are along with the increase of fault severity level, information entropy parameter increases gradually, when information entropy parameter is greater than 1.5, represents that fault degree is more serious; For quality imbalance fault, its fault degree is more serious, and information entropy parameter is less, when information entropy parameter is less than 0.3, characterizes fault more serious.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510539779.9A CN105157821A (en) | 2015-08-30 | 2015-08-30 | Rotary machinery vibration fault diagnosis and quantitative analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510539779.9A CN105157821A (en) | 2015-08-30 | 2015-08-30 | Rotary machinery vibration fault diagnosis and quantitative analysis method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105157821A true CN105157821A (en) | 2015-12-16 |
Family
ID=54798760
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510539779.9A Pending CN105157821A (en) | 2015-08-30 | 2015-08-30 | Rotary machinery vibration fault diagnosis and quantitative analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105157821A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106595850A (en) * | 2016-12-21 | 2017-04-26 | 潘小胜 | Mechanical oscillation signal fault analysis method |
| CN106840379A (en) * | 2017-03-08 | 2017-06-13 | 潘小胜 | A kind of failure analysis methods of mechanical oscillation signal |
| 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 |
| CN109374119A (en) * | 2018-09-29 | 2019-02-22 | 国网山西省电力公司阳泉供电公司 | Transformer vibration signal Characteristic Extraction method |
| CN112082720A (en) * | 2020-09-04 | 2020-12-15 | 江苏方天电力技术有限公司 | Method for determining early warning value of vibration fault of fixed-speed rotating machine |
| CN113639945A (en) * | 2021-06-28 | 2021-11-12 | 上海宇航***工程研究所 | Spacecraft random vibration test condition design method based on empirical mode decomposition |
| WO2022261805A1 (en) * | 2021-06-15 | 2022-12-22 | 大连理工大学 | Diesel engine gearbox fault diagnosis method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09210766A (en) * | 1996-01-31 | 1997-08-15 | Toshiba Corp | Vibration measuring apparatus |
| CN102175307A (en) * | 2011-02-01 | 2011-09-07 | 华北电力大学 | Real-time quantitative analysis method for low-frequency vibration spectrum array of steam turbine generator set |
| CN102288280A (en) * | 2011-05-09 | 2011-12-21 | 华北电力大学（保定） | Shafting torsional vibration extraction method |
-
2015
- 2015-08-30 CN CN201510539779.9A patent/CN105157821A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09210766A (en) * | 1996-01-31 | 1997-08-15 | Toshiba Corp | Vibration measuring apparatus |
| CN102175307A (en) * | 2011-02-01 | 2011-09-07 | 华北电力大学 | Real-time quantitative analysis method for low-frequency vibration spectrum array of steam turbine generator set |
| CN102288280A (en) * | 2011-05-09 | 2011-12-21 | 华北电力大学（保定） | Shafting torsional vibration extraction method |
Non-Patent Citations (1)
| Title |
|---|
| 邴汉昆: "基于小波分析和SVM的汽轮机非线性振动故障诊断研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106595850A (en) * | 2016-12-21 | 2017-04-26 | 潘小胜 | Mechanical oscillation signal fault analysis method |
| CN106840379A (en) * | 2017-03-08 | 2017-06-13 | 潘小胜 | A kind of failure analysis methods of mechanical oscillation signal |
| 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 |
| CN109374119A (en) * | 2018-09-29 | 2019-02-22 | 国网山西省电力公司阳泉供电公司 | Transformer vibration signal Characteristic Extraction method |
| CN112082720A (en) * | 2020-09-04 | 2020-12-15 | 江苏方天电力技术有限公司 | Method for determining early warning value of vibration fault of fixed-speed rotating machine |
| WO2022261805A1 (en) * | 2021-06-15 | 2022-12-22 | 大连理工大学 | Diesel engine gearbox fault diagnosis method |
| CN113639945A (en) * | 2021-06-28 | 2021-11-12 | 上海宇航***工程研究所 | Spacecraft random vibration test condition design method based on empirical mode decomposition |
| CN113639945B (en) * | 2021-06-28 | 2024-02-09 | 上海宇航***工程研究所 | Spacecraft random vibration test condition design method based on empirical mode decomposition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105157821A (en) | Rotary machinery vibration fault diagnosis and quantitative analysis method | |
| CN108343599B (en) | A kind of water pump assembly intelligent failure diagnosis method based on more granularities cascade forest | |
| CN101907437B (en) | Wavelet difference algorithm-based cable fault localization method | |
| CN103064008B (en) | A kind of Nolinear analog circuit soft fault diagnostic method based on Hilbert-Huang transform | |
| CN105092239B (en) | A kind of initial failure of gear detection method | |
| CN104165925B (en) | The centrifugal compressor half-opened impeller crack fault detection method of accidental resonance | |
| CN104330471B (en) | Lamb wave time-varying probability model monitoring method for aviation structure damage | |
| Zhou et al. | Application of the horizontal slice of cyclic bispectrum in rolling element bearings diagnosis | |
| CN115688018B (en) | Method for monitoring state and diagnosing faults of bearing under multiple working conditions | |
| CN105758644A (en) | Rolling bearing fault diagnosis method based on variation mode decomposition and permutation entropy | |
| CN109102032A (en) | A kind of pumping plant unit diagnostic method based on depth forest and oneself coding | |
| Song et al. | Self-adaptive multivariate variational mode decomposition and its application for bearing fault diagnosis | |
| CN106447039A (en) | Non-supervision feature extraction method based on self-coding neural network | |
| CN105138770B (en) | Space product Reliablility simulation appraisal procedure based on indirect characteristic quantities | |
| Sheng et al. | Applications in bearing fault diagnosis of an improved Kurtogram algorithm based on flexible frequency slice wavelet transform filter bank | |
| CN107036808B (en) | Gearbox of wind turbine combined failure diagnostic method based on support vector machines probability Estimation | |
| CN105547730A (en) | Fault detection system of water-wheel generator set | |
| CN109241915A (en) | The intelligent power plant's pump machine method for diagnosing faults screened based on the steady non-stationary differentiation of vibration signal and feature | |
| CN112326213A (en) | Abnormal data detection method and device and mechanical fault detection method and device | |
| CN105352726B (en) | A kind of method for diagnosing faults of gear | |
| CN105510066A (en) | Adaptive-noise-reduction-algorithm-based same-class rotary machinery system fault diagnosis method | |
| CN103699117A (en) | Method and system for diagnosing failure based on actual working conditions of nuclear power plant and simulation system | |
| CN105258790A (en) | Field vibration signal de-noising parameter optimization selection method | |
| CN107559228B (en) | Method based on bispectral data detection and diagnosis fan trouble | |
| CN106596110A (en) | Online data-based hydroelectric generating set hydraulic unbalance fault automatic analysis and diagnosis 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151216 |
|
| RJ01 | Rejection of invention patent application after publication |