CN114414242A - Large-scale rotating equipment spindle state signal feature extraction method based on multi-domain analysis and Relieff - Google Patents
Large-scale rotating equipment spindle state signal feature extraction method based on multi-domain analysis and Relieff Download PDFInfo
- Publication number
- CN114414242A CN114414242A CN202111556632.2A CN202111556632A CN114414242A CN 114414242 A CN114414242 A CN 114414242A CN 202111556632 A CN202111556632 A CN 202111556632A CN 114414242 A CN114414242 A CN 114414242A
- Authority
- CN
- China
- Prior art keywords
- relieff
- sample
- extracting
- domain
- main shaft
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 21
- 238000000605 extraction Methods 0.000 title claims abstract description 8
- 238000000034 method Methods 0.000 claims abstract description 33
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 4
- 238000004364 calculation method Methods 0.000 claims description 14
- 238000003745 diagnosis Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 3
- 238000000926 separation method Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 abstract description 2
- 238000003058 natural language processing Methods 0.000 abstract 1
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000014509 gene expression Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/02—Gearings; Transmission mechanisms
- G01M13/028—Acoustic or vibration analysis
Landscapes
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a method for extracting the main shaft state signal characteristics of large-scale rotating equipment based on multi-domain analysis and Relieff. The invention relates to the technical field of natural language processing, and the invention extracts time domain characteristics of signals based on vibration signals, wherein the time domain characteristics comprise dimensional parameters and dimensionless characteristic parameters; extracting frequency domain characteristics including center-of-gravity frequency, average frequency, root-mean-square frequency and frequency standard deviation by adopting Fourier transform based on the extracted time domain characteristics of the vibration signal; decomposing each sub-band of the vibration signal based on wavelet packet decomposition, and adopting each sub-band to complete energy ratio of each sub-band so as to extract time-frequency domain characteristic indexes; calculating the weight of the multi-domain feature based on the Relieff, and finishing the dimensionality reduction processing of the multi-domain feature index according to the weight; and irrelevant indexes are removed based on the weight values of the indexes, so that the accurate extraction of the main shaft state signal characteristics of the large-scale rotary equipment is realized.
Description
Technical Field
The invention relates to the technical field of fault diagnosis in a main shaft of large-scale rotating equipment and large-scale rotating equipment, in particular to a method for extracting state signal characteristics of the main shaft of the large-scale rotating equipment based on multi-domain analysis and Relieff.
Background
The numerical control machine tool is an important mark for measuring the development level of the equipment manufacturing industry as a 'working master machine' of the equipment manufacturing industry, is a foundation for generating the value of the manufacturing industry and a fulcrum of industry leap, and is the core of the basic manufacturing capacity.
The investigation shows that the annual maintenance cost of the numerical control large-scale rotating equipment in China reaches more than 10% of the total cost of the numerical control large-scale rotating equipment, the faults are frequently generated by 30% in the main problems of the numerical control large-scale rotating equipment in China, and the faults of a main shaft system account for about 57% of the total faults of the large-scale rotating equipment, so that effective fault diagnosis is required to be carried out, and the normal work of the numerical control large-scale rotating equipment is ensured.
When a main shaft of large-scale rotating equipment breaks down, vibration signals of a system obviously change, but how to effectively extract state features from a large number of main shaft state signals is a core element.
Disclosure of Invention
The invention provides a method for extracting main shaft state signal features of large-scale rotating equipment based on multi-domain analysis and Relieff, which is used for respectively triggering from multiple domains of signals based on main shaft state signals and carrying out signal feature calculation and extraction by taking time domain, frequency domain and time frequency domain as representatives. The invention provides the following technical scheme:
a method for extracting main shaft state signal features of large-scale rotating equipment based on multi-domain analysis and Relieff comprises the following steps:
step 1: extracting time domain characteristics of the signals based on the vibration signals, wherein the time domain characteristics comprise dimensional parameters and dimensionless characteristic parameters;
step 2: extracting frequency domain characteristics including center-of-gravity frequency, average frequency, root-mean-square frequency and frequency standard deviation by adopting Fourier transform based on the extracted time domain characteristics of the vibration signal;
and step 3: decomposing each sub-band of the vibration signal based on wavelet packet decomposition, and adopting each sub-band to complete energy ratio of each sub-band so as to extract time-frequency domain characteristic indexes;
and 4, step 4: calculating the weight of the multi-domain feature based on the Relieff, and finishing the dimensionality reduction processing of the multi-domain feature index according to the weight;
and 5: and irrelevant indexes are removed based on the weight values of the indexes, so that the accurate extraction of the main shaft state signal characteristics of the large-scale rotary equipment is realized.
Preferably, the dimensional parameters include signal peak-to-peak, maximum-to-minimum, and average amplitude; the dimensionless parameters include a form factor, a kurtosis factor, and a pulse factor.
Preferably, the step 3 specifically comprises:
extracting the time-frequency domain characteristics of signals by adopting wavelet packet separation, and selecting reasonable decomposition layer number n and wavelet basis functions, wherein n is determined by the formula (1):
the selection of the wavelet basis functions directly influences the decomposition effect, the wavelet basis functions comprise db series, sym series and coif series, then the frequency band energy is solved, and finally the energy proportion of each frequency band is solved to obtain the time-frequency domain signal feature vector.
Preferably, db series wavelet basis functions, in particular db5 wavelet, are used as decomposition basis functions for the vibration signals of the spindle system.
Preferably, the vibration signals of the different state phases of the bearing are decomposed, and the number of the decomposed layers is 3.
Preferably, the step 4 specifically includes:
the dimensionality reduction of multi-domain characteristic indexes is completed based on a Relieff algorithm, the multi-index selection problem is sequentially converted into two types of index selection, a sample class set is set to be D, an original characteristic index set is in a lambda class, and fiFor each class of features, the sample set N ═ f1,f2,……fλAnd (4) setting R as a sample in the sample set N, respectively calculating the distances between the sample R and the similar and the heterogeneous samples, selecting k similar and heterogeneous neighbor samples according to the distances, respectively marking the similar and the heterogeneous neighbor samples as T and Y, and calculating and updating the weight of each characteristic index according to the similar and the heterogeneous neighbor samples.
Preferably, when the weight is updated, a certain feature x in the sample R is selected, the distance to the similar neighbor sample and the distance to the heterogeneous neighbor sample are sequentially calculated, and the update weight is compared, where the weight is as shown in formula (2):
where m is the number of sampling times, p (C) is the proportion of the sample class C in the total sample set N, diff (x, R, T)i) Representing a characteristic sample R and a similar neighbor sample TiThe difference value on the characteristic index x is calculated in a Euclidean distance between objects, and calculation formulas when x is divided into discrete and continuous are respectively shown as formulas (3) and (4):
preferably, the cycle count N is set to 30, the sampling count m is set to 100, and the number k of neighboring samples is set to 50, and after the cycle count is calculated, the average value of the feature weights is obtained.
Preferably, when the euclidean distance between the sample R and the similar sample is smaller and the distance between the heterogeneous samples is larger, the weight w is also increased, which conforms to the rule of the mutual close of the similar samples and the mutual principle of the heterogeneous samples.
Preferably, the step 5 specifically comprises: and removing the indexes with smaller weights, reducing the dimensionality of the extracted indexes, reducing subsequent calculation amount and ensuring the fault diagnosis accuracy.
The invention has the following beneficial effects:
according to the method for extracting the state signal characteristics of the main shaft of the large-scale rotating equipment based on the multi-domain analysis and the Relieff algorithm, the state indexes of a large number of original state signals of the main shaft of the large-scale rotating equipment are extracted, the redundancy and the complexity of the original signals are avoided, the state characteristics of the main shaft of the large-scale rotating equipment are fully reserved through the extraction of the characteristics, the dimension reduction of the experimental characteristic indexes is further performed based on the Relieff algorithm, the calculated amount is further reduced, and data support is provided for subsequent fault diagnosis.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a time domain characteristic index of a vibration signal;
FIG. 3 is a frequency domain indicator of a vibration signal;
FIG. 4 is a schematic diagram of wavelet packet decomposition;
FIG. 5 is a diagram of energy ratios of sub-bands of wavelet packet decomposition;
fig. 6 is a graph of the ratio of each index weight based on the ReliefF algorithm.
Detailed Description
The present invention will be described in detail with reference to specific examples.
The first embodiment is as follows:
as shown in fig. 1 to 6, a specific optimization technical solution adopted to solve the above technical problems is: the invention relates to a method for extracting the state signal characteristics of a main shaft of large-scale rotating equipment based on multi-domain analysis and Relieff, which comprises the following steps:
a method for extracting main shaft state signal features of large-scale rotating equipment based on multi-domain analysis and Relieff comprises the following steps:
step 1: extracting time domain characteristics of the signals based on the vibration signals, wherein the time domain characteristics comprise dimensional parameters and dimensionless characteristic parameters; the dimensional parameters comprise signal peak value, maximum minimum value and average amplitude; the dimensionless parameters include form factor, kurtosis factor, and impulse factor.
The method comprises the following steps: firstly, the time domain characteristics of the signal are extracted based on the vibration signal, and the calculation formula of each time domain index is shown in table 1.
TABLE 1 time-domain index calculation expression
The time domain index of the bearing in different states is shown in figure 2.
Step 2: extracting frequency domain characteristics including center-of-gravity frequency, average frequency, root-mean-square frequency and frequency standard deviation by adopting Fourier transform based on the extracted time domain characteristics of the vibration signal;
typical frequency domain characteristics are shown in table 2.
TABLE 2 frequency domain index calculation expressions
The frequency domain characteristics of the bearings in different states are calculated as shown in fig. 3.
And step 3: decomposing each sub-band of the vibration signal based on wavelet packet decomposition, and adopting each sub-band to complete energy ratio of each sub-band so as to extract time-frequency domain characteristic indexes;
the step 3 specifically comprises the following steps:
the schematic diagram of the wavelet packet decomposition principle is shown in fig. 4. The method comprises the following steps of extracting signal time-frequency domain characteristics by wavelet packet separation, and selecting reasonable decomposition layer number n and wavelet basis functions, wherein n is determined by formula (1):
the selection of the wavelet basis functions directly influences the decomposition effect, the wavelet basis functions comprise db series, sym series and coif series, then the frequency band energy is solved, and finally the energy proportion of each frequency band is solved to obtain the time-frequency domain signal feature vector.
And a db series wavelet basis function is adopted for the vibration signal of the main shaft system, and specifically a db5 wavelet is adopted as a decomposition basis function. The vibration signals of the bearings in different phases are decomposed, and the number of the decomposed layers is 3 as shown in fig. 5.
And 4, step 4: calculating the weight of the multi-domain feature based on the Relieff, and finishing the dimensionality reduction processing of the multi-domain feature index according to the weight;
the step 4 specifically comprises the following steps:
the method for completing the dimensionality reduction of the multi-domain characteristic index based on the Relieff algorithmThe method converts the multi-index selection problem into two types of index selection in turn, and overcomes the defect that only two types of index selection can be carried out. Sequentially converting the multi-index selection problem into two types of index selection, setting a sample classification set as D, and setting an original characteristic index set to be in a lambda type and fiFor each class of features, the sample set N ═ f1,f2,……fλAnd (4) setting R as a sample in the sample set N, respectively calculating the distance between the sample R and the same type and the different type, selecting k similar and different type neighbor samples according to the distance, respectively marking the similar and different type neighbor samples as T and Y, and calculating and updating the weight of each characteristic index according to the similar and different type neighbor samples.
When the weight is updated, selecting a certain characteristic x in the sample R, sequentially calculating the distance between the sample R and the similar neighbor sample and the distance between the sample R and the heterogeneous neighbor sample, and comparing and updating the weight, wherein the weight is shown as a formula (2):
where m is the number of sampling times, p (C) is the proportion of the sample class C in the total sample set N, diff (x, R, T)i) Representing a characteristic sample R and a similar neighbor sample TiThe difference value on the characteristic index x is calculated in a Euclidean distance between objects, and calculation formulas when x is divided into discrete and continuous are respectively shown as formulas (3) and (4):
and setting the cycle calculation frequency N as 30, the sampling frequency m as 100, the number k of adjacent samples as 50, and calculating the average value of each characteristic weight after the cycle calculation.
Preferably, when the euclidean distance between the sample R and the similar sample is smaller and the distance between the heterogeneous samples is larger, the weight w is also increased, which conforms to the rule of the mutual close of the similar samples and the mutual principle of the heterogeneous samples.
And 5: and irrelevant indexes are removed based on the weight values of the indexes, so that the accurate extraction of the main shaft state signal characteristics of the large-scale rotary equipment is realized.
The step 5 specifically comprises the following steps: and removing the indexes with smaller weights, reducing the dimensionality of the extracted indexes, reducing subsequent calculation amount and ensuring the fault diagnosis accuracy.
The above is only a preferred embodiment of the method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the ReliefF, and the protection range of the method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the ReliefF is not limited to the above embodiments, and all technical solutions belonging to the thought belong to the protection range of the invention. It should be noted that modifications and variations which do not depart from the gist of the invention are intended to be within the scope of the invention.
Claims (10)
1. A method for extracting the main shaft state signal characteristics of large-scale rotating equipment based on multi-domain analysis and Relieff is characterized by comprising the following steps: the method comprises the following steps:
step 1: extracting time domain characteristics of the signals based on the vibration signals, wherein the time domain characteristics comprise dimensional parameters and dimensionless characteristic parameters;
step 2: extracting frequency domain characteristics including center-of-gravity frequency, average frequency, root-mean-square frequency and frequency standard deviation by adopting Fourier transform based on the extracted time domain characteristics of the vibration signal;
and step 3: decomposing each sub-band of the vibration signal based on wavelet packet decomposition, and adopting each sub-band to complete energy ratio of each sub-band so as to extract time-frequency domain characteristic indexes;
and 4, step 4: calculating the weight of the multi-domain feature based on the Relieff, and finishing the dimensionality reduction processing of the multi-domain feature index according to the weight;
and 5: and irrelevant indexes are removed based on the weight values of the indexes, so that the accurate extraction of the main shaft state signal characteristics of the large-scale rotary equipment is realized.
2. The method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the Relieff as claimed in claim 1, wherein the method comprises the following steps: the dimensional parameters comprise signal peak value, maximum minimum value and average amplitude; the dimensionless parameters include form factor, kurtosis factor, and impulse factor.
3. The method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the Relieff as claimed in claim 2, wherein the method comprises the following steps: the step 3 specifically comprises the following steps:
extracting the time-frequency domain characteristics of signals by adopting wavelet packet separation, and selecting reasonable decomposition layer number n and wavelet basis functions, wherein n is determined by the formula (1):
the selection of the wavelet basis functions directly influences the decomposition effect, the wavelet basis functions comprise db series, sym series and coif series, then the frequency band energy is solved, and finally the energy proportion of each frequency band is solved to obtain the time-frequency domain signal feature vector.
4. The method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the Relieff as claimed in claim 3, wherein the method comprises the following steps: and db series wavelet basis functions are adopted for the vibration signals of the main shaft system, and db5 wavelets are specifically adopted as decomposition basis functions.
5. The method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the Relieff as claimed in claim 4, wherein the method comprises the following steps: and (3) decomposing vibration signals of the bearing in different state phases, wherein the number of the decomposition layers is 3.
6. The method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the Relieff as claimed in claim 5, wherein the method comprises the following steps: the step 4 specifically comprises the following steps:
finishing the dimensionality reduction of multi-domain characteristic indexes based on a Relieff algorithm, sequentially converting the multi-index selection problem into two types of index selection, setting a sample classification set as D, wherein an original characteristic index set is in a lambda class and fiFor each class of features, the sample set N ═ f1,f2,……fλAnd (4) setting R as a sample in the sample set N, respectively calculating the distances between the sample R and the similar and the heterogeneous samples, selecting k similar and heterogeneous neighbor samples according to the distances, respectively marking the similar and the heterogeneous neighbor samples as T and Y, and calculating and updating the weight of each characteristic index according to the similar and the heterogeneous neighbor samples.
7. The method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the Relieff as claimed in claim 6, wherein the method comprises the following steps: when the weight is updated, selecting a certain characteristic x in the sample R, sequentially calculating the distance between the sample R and the similar neighbor sample and the distance between the sample R and the heterogeneous neighbor sample, and comparing and updating the weight, wherein the weight is shown as a formula (2):
where m is the number of sampling times, p (C) is the proportion of the sample class C in the total sample set N, diff (x, R, T)i) Representing a characteristic sample R and a similar neighbor sample TiThe difference value on the characteristic index x is calculated in a Euclidean distance between objects, and calculation formulas when x is divided into discrete and continuous are respectively shown as formulas (3) and (4):
8. the method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the Relieff as claimed in claim 7, wherein the method comprises the following steps: and setting the cycle calculation number N to be 30, the sampling number m to be 100 and the number k of adjacent samples to be 50, and calculating the average value of the feature weights after the cycle calculation.
9. The method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the Relieff as claimed in claim 8, wherein the method comprises the following steps:
when the Euclidean distance between the sample R and the similar sample is smaller and the distance between the different samples is larger, the weight w is increased, and the rule that the similar samples are close to each other and the different samples are mutually principle is met.
10. The method for extracting the main shaft state signal feature of the large-scale rotating equipment based on the multi-domain analysis and the Relieff as claimed in claim 9, wherein the method comprises the following steps: the step 5 specifically comprises the following steps: and removing the indexes with smaller weights, reducing the dimensionality of the extracted indexes, reducing subsequent calculation amount and ensuring the fault diagnosis accuracy.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111556632.2A CN114414242A (en) | 2021-12-17 | 2021-12-17 | Large-scale rotating equipment spindle state signal feature extraction method based on multi-domain analysis and Relieff |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111556632.2A CN114414242A (en) | 2021-12-17 | 2021-12-17 | Large-scale rotating equipment spindle state signal feature extraction method based on multi-domain analysis and Relieff |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114414242A true CN114414242A (en) | 2022-04-29 |
Family
ID=81268015
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111556632.2A Pending CN114414242A (en) | 2021-12-17 | 2021-12-17 | Large-scale rotating equipment spindle state signal feature extraction method based on multi-domain analysis and Relieff |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114414242A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104655423A (en) * | 2013-11-19 | 2015-05-27 | 北京交通大学 | Rolling bearing fault diagnosis method based on time-frequency domain multidimensional vibration feature fusion |
| CN112347588A (en) * | 2020-11-26 | 2021-02-09 | 中国舰船研究设计中心 | Rotary machine fault diagnosis method based on wavelet packet decomposition |
| CN112906473A (en) * | 2021-01-19 | 2021-06-04 | 杭州安脉盛智能技术有限公司 | Fault diagnosis method for rotating equipment |
-
2021
- 2021-12-17 CN CN202111556632.2A patent/CN114414242A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104655423A (en) * | 2013-11-19 | 2015-05-27 | 北京交通大学 | Rolling bearing fault diagnosis method based on time-frequency domain multidimensional vibration feature fusion |
| CN112347588A (en) * | 2020-11-26 | 2021-02-09 | 中国舰船研究设计中心 | Rotary machine fault diagnosis method based on wavelet packet decomposition |
| CN112906473A (en) * | 2021-01-19 | 2021-06-04 | 杭州安脉盛智能技术有限公司 | Fault diagnosis method for rotating equipment |
Non-Patent Citations (3)
| Title |
|---|
| XIAOMING WANG 等: "A Health Monitoring Method of Machine Tool Spindle Based on Multi-domain Analysis and Convolutional Neural Network", 《JOURNAL OF PHYSICS: CONFERENCE SERIES》, 19 April 2021 (2021-04-19), pages 1 - 5 * |
| 吴勇: "基于振动信号的滚动轴承故障诊断技术研究", 《中国优秀硕士论文工程科技Ⅱ辑》, 15 February 2020 (2020-02-15), pages 14 - 24 * |
| 王友荣: "Relief F加权特征选择方法在旋转机械故障诊断中的应用研究", 《中国优秀硕士论文工程科技II辑》, 15 May 2015 (2015-05-15), pages 12 - 16 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104655425B (en) | Bearing fault classification diagnosis method based on sparse representation and LDM (large margin distribution machine) | |
| CN111089726B (en) | Rolling bearing fault diagnosis method based on optimal dimension singular spectrum decomposition | |
| CN110427916B (en) | Fault diagnosis method under action of central frequency convergence tendency | |
| AU2021105779A4 (en) | Planetary Gearbox Fault Diagnosis Method Using Parameter Optimized VMD and Multi-domain Manifold Learning | |
| CN112257530B (en) | Rolling bearing fault diagnosis method based on blind signal separation and support vector machine | |
| CN103345200B (en) | A kind of cut Identification of Chatter method based on generalized interval | |
| CN105196114A (en) | Real-time online tool wear monitoring method based on wavelet analysis and neural network | |
| CN104634571A (en) | Fault diagnosis method for rolling bearing based on LCD-MF (Local Characteristic Scale Decomposition )-(Multifractal) | |
| CN106468751A (en) | A kind of transformer winding state recognition methodss of the neutral net that resonated based on fuzzy self-adaption | |
| CN109827777A (en) | Rolling bearing fault prediction technique based on Partial Least Squares extreme learning machine | |
| CN113177537B (en) | Fault diagnosis method and system for rotary mechanical equipment | |
| CN113420691A (en) | Mixed domain characteristic bearing fault diagnosis method based on Pearson correlation coefficient | |
| CN111337256A (en) | Method for diagnosing fault depth local migration of rolling bearing weighted by domain asymmetry factor | |
| CN113378725A (en) | Cutter fault diagnosis method, equipment and storage medium based on multi-scale-channel attention network | |
| CN111665050B (en) | Rolling bearing fault diagnosis method based on clustering K-SVD algorithm | |
| CN104778348B (en) | Voltage sag action characteristic modeling method for low-voltage trip | |
| CN112364706A (en) | Small sample bearing fault diagnosis method based on class imbalance | |
| CN112633098A (en) | Fault diagnosis method and system for rotary machine and storage medium | |
| CN114462480A (en) | Multi-source sensor rolling mill fault diagnosis method based on non-equilibrium data set | |
| CN114414242A (en) | Large-scale rotating equipment spindle state signal feature extraction method based on multi-domain analysis and Relieff | |
| Yu et al. | Rolling bearing fault feature extraction and diagnosis method based on MODWPT and DBN | |
| CN111382792B (en) | Rolling bearing fault diagnosis method based on double-sparse dictionary sparse representation | |
| CN102778632A (en) | Double normalization recognition method for directly forecasting and recognizing transformer winding fault type | |
| CN105116323B (en) | A kind of electrical fault detection method based on RBF | |
| CN109426216B (en) | Rectangular plane machining depth identification method for numerical control milling machine |
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 |