CN110749443A - Rolling bearing fault diagnosis method and system based on high-order origin moment - Google Patents
Rolling bearing fault diagnosis method and system based on high-order origin moment Download PDFInfo
- Publication number
- CN110749443A CN110749443A CN201911181169.0A CN201911181169A CN110749443A CN 110749443 A CN110749443 A CN 110749443A CN 201911181169 A CN201911181169 A CN 201911181169A CN 110749443 A CN110749443 A CN 110749443A
- Authority
- CN
- China
- Prior art keywords
- vector
- data
- rolling bearing
- fault
- calculating
- 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
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
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
Abstract
The application discloses a rolling bearing fault diagnosis method and system based on a high-order origin moment, wherein sample data of normal working conditions and fault working conditions in the operation of a rolling bearing are extracted, and the data are subjected to standardized processing; dividing the time domain signal into five layers of signals by adopting a wavelet decomposition method; calculating four-order origin moments of a normal working condition and three fault working conditions; combining the four-order origin moments of the five-layer signals obtained by calculation into two vectors which are a vector A and a vector B respectively by taking the angle of the included angle of the vectors as a characteristic, and calculating the included angle between the vector A and the vector B; determining the distinguishing ranges of the four working conditions, and calculating the accuracy of different data volumes; establishing an index function of the relation between the accuracy and the data volume, and selecting the optimal data volume; and establishing an SVM (support vector machine) classifier, and respectively inputting the training sample and the test sample into the SVM classifier for classification diagnosis. And then the characteristics of the rolling bearing faults can be effectively extracted, and the identification degree of the rolling bearing faults is improved.
Description
Technical Field
The application relates to the technical field of rolling bearing fault diagnosis, in particular to a rolling bearing fault diagnosis method and system based on a high-order origin moment.
Background
The dispersion of the life of the rolling bearing is larger than that of other rotating mechanical devices, and because the rolling bearing has the characteristics, some rolling bearings have various faults when the design service life of the rolling bearing is not reached. Therefore, the operation data of the rolling bearing is collected, and the working state and the fault of the rolling bearing are diagnosed and predicted in real time based on the collected data, so that whether the equipment can safely, reliably and stably operate or not is judged, and the timely maintenance and treatment of the faulted bearing can effectively avoid the influence. Therefore, it is necessary to collect the original vibration signals of the rolling bearing failure.
At present, the rolling bearing is widely applied to rotary machinery, and the running state of the rolling bearing often directly influences the precision, the reliability and the service life of the whole machine. The service life of the rolling bearing is very discrete, and the rolling bearing cannot be maintained at regular time, so that the rolling bearing monitoring and fault diagnosis method has important significance for the state monitoring and fault diagnosis of the rolling bearing.
In the prior art, most of fault diagnosis and analysis of the rolling bearing are based on the acquired signal attributes, however, most of the signals have the characteristics of nonlinearity, non-stability and the like, and the information of the state of the rolling bearing can be obtained through analysis. The rolling bearing fault acquisition method has the advantages that the identification degree of faults is low due to the complex conditions of nonlinearity, non-stability and the like of signal attributes of the rolling bearing fault acquisition.
Disclosure of Invention
In order to solve the technical problems, the following technical scheme is provided:
in a first aspect, an embodiment of the present application provides a rolling bearing fault diagnosis method based on a high-order origin moment, where the method includes: extracting sample data of normal working conditions and fault working conditions in the operation of the rolling bearing, and carrying out standardized processing on the data; dividing the time domain signal into five layers of signals by adopting a wavelet decomposition method; calculating four-order origin moments of a normal working condition and three fault working conditions; combining the four-order origin moments (1,2,3,4,5) of the five-layer signals obtained by calculation into two vectors, namely a vector A (1,2,3) and a vector B (3,4,5), by taking a vector included angle as a characteristic, and calculating an included angle between the vector A (1,2,3) and the vector B (3,4, 5); determining the distinguishing ranges of the four working conditions, and calculating the accuracy of different data volumes; establishing an index function of the relation between the accuracy and the data volume, and selecting the optimal data volume; and establishing an SVM (support vector machine) classifier, and respectively inputting the training sample and the test sample into the SVM classifier for classification diagnosis.
By adopting the implementation mode, the processed data are classified through processing the sample data of the normal working condition and the fault working condition of the rolling bearing, and the classified diagnosis is carried out on the data of different types, so that the characteristics of the faults of the rolling bearing can be effectively extracted, and the identification degree of the faults of the rolling bearing is improved.
With reference to the first aspect, in a first possible implementation manner of the first aspect, the extracting sample data of normal operating conditions and fault operating conditions in operation of the rolling bearing, and performing normalization processing on the data includes:
wherein x is*The data is normalized; x is original data; x is the number ofmaxIs the maximum value in the original data; x is the number ofminIs the minimum value in the raw data.
Extracting sample data of normal working conditions and fault working conditions of the rolling bearing, respectively carrying out standardization processing on the sample data of the normal working conditions and the fault working conditions of the rolling bearing, and drawing time domain waveforms; the standardization processing is mainly proposed for the convenience of data processing, large data can be mapped into a small range for processing, and small data can be processed in a large range, so that the standardization processing is more convenient and faster.
With reference to the first aspect, in a second possible implementation manner of the first aspect, the dividing the time-domain signal into five layers of signals by using a wavelet decomposition method includes: a wavelet decomposition method is adopted to select the db8 wavelet in the time domain signals as a wavelet base, and the db8 wavelet is divided into five layers of signals.
With reference to the first aspect, in a third possible implementation manner of the first aspect, the calculating fourth-order origin moments of the normal operating condition and the three fault operating conditions includes:
wherein k is the fourth order origin moment, a is data, and n is the data size.
With reference to the first aspect, in a fourth possible implementation manner of the first aspect, the combining the calculated four-order origin moments (1,2,3,4,5) of the five-layer signals into two vectors, namely vector a (1,2,3) and vector B (3,4,5), and calculating an angle between vector a (1,2,3) and vector B (3,4,5) by using a vector included angle as a feature includes:
the vector included angle is characterized in that five layers of signals obtained by wavelet decomposition are calculated to obtain a fourth order origin moment, five layers of signals obtain five fourth order origin moments, and the five layers of signals are perpendicular to each other in space, so that the fourth order origin moments calculated by the signals of each layer are combined into a group of space vectors;
sequentially arranging to obtain a group of vectors (1,2,3,4,5), combining the vectors (1,2,3,4,5) into two vectors, namely a vector A (1,2,3) and a vector B (3,4,5), and calculating an included angle between the vector A (1,2,3) and the vector B (3,4, 5); the formula is as follows:
wherein theta represents an angle of rotation of the rotary shaft,
represents the vectors (1,2,3),representing the vector (3,4, 5).
With reference to the first aspect, in a fifth possible implementation manner of the first aspect, the determining the range of the four operating conditions, and calculating the accuracy of different data sizes includes: calculating four-order origin moments of four working conditions; calculating a characteristic angle of each working condition; and taking the characteristic angle obtained by calculating the training data as a training characteristic, and then calculating the average value of the training angle of each working condition to obtain four average angles of the four working conditions, namely the average angle of normal operation, the average angle of the inner ring fault, the average angle of the ball fault and the average angle of the outer ring fault in sequence.
After averaging again between two average angles of two adjacent working conditions, a distinguishing value can be calculated and used as the upper limit and the lower limit of a distinguishing range of the two adjacent working conditions; then, determining the ranges of the four working conditions according to the distinguishing values, wherein the range of the normal working conditions is that theta is more than or equal to 54 and less than 100; the failure range of the inner ring is more than or equal to 19 and less than 54; the failure range of the ball is more than or equal to 9 and less than 19; the outer ring fault range is more than or equal to 0 and less than 9. After the range is determined, distinguishing the characteristic angles calculated in the training data and calculating the training accuracy; when the four working conditions are all in the corresponding ranges, the condition is judged to be correct. If one of the operating conditions is not within the corresponding range, it is determined that the condition is erroneous.
With reference to the first aspect, in a sixth possible implementation manner of the first aspect, the establishing an indicator function of a relationship between a correct rate and a data amount includes:
where c is the accuracy, d is the data size, α is the adjustment factor, and y is the index function.
When the index function image has a unique highest point, the adjustment coefficient α is determined to be 0.6, the highest point is the function optimal point, the corresponding data volume is determined to be the optimal data volume, the value range of the adjustment coefficient α is (α is 0,0.05,0.1, …,1), and the value range of the data volume d is (d is 1000,2000, …, 10000).
With reference to the first aspect, in a seventh possible implementation manner of the first aspect, the establishing an SVM support vector machine classifier, and inputting the training samples and the test samples into the SVM classifier respectively for classification diagnosis includes: establishing an SVM (support vector machine) classifier, and inputting the training sample and the test sample into the SVM classifier respectively for classification diagnosis; and the identification result 1 represents a normal working condition, 2 represents an inner ring fault, 3 represents a ball fault, and 4 represents an outer ring fault, so that the fault diagnosis of the differentiation of the test samples is realized.
In a second aspect, the present application provides a rolling bearing fault diagnosis system based on a high-order origin moment, where the system includes: the extraction module is used for extracting sample data of normal working conditions and fault working conditions in the operation of the rolling bearing and carrying out standardized processing on the data; the wavelet decomposition module is used for dividing the time domain signals into five layers of signals by adopting a wavelet decomposition method; the first calculation module is used for calculating the fourth-order origin moments of the normal working condition and the three fault working conditions; the second calculation module is used for combining the calculated five-layer signals into two vectors, namely a vector A (1,2,3) and a vector B (3,4,5), by taking the included angle of the vectors as a characteristic, and calculating the included angle between the vector A (1,2,3) and the vector B (3,4, 5); the determining module is used for determining the distinguishing ranges of the four working conditions and calculating the accuracy of different data volumes; the establishing module is used for establishing an index function of the relation between the accuracy and the data volume and selecting the optimal data volume; and the fault diagnosis module is used for establishing an SVM (support vector machine) classifier and inputting the training samples and the test samples into the SVM classifier respectively for classification diagnosis.
Drawings
Fig. 1 is a schematic flowchart of a rolling bearing fault diagnosis method based on a high-order origin moment according to an embodiment of the present application;
FIG. 2 is a time domain diagram of normal operating conditions and fault operating conditions of a rolling bearing provided by the embodiment of the application;
FIG. 3 is a five-level signal diagram of normal signal wavelet decomposition provided by an embodiment of the present application;
fig. 4 is a five-layer signal diagram of wavelet decomposition of a fault of an inner ring of a rolling bearing provided by an embodiment of the application;
FIG. 5 is a five-layer signal diagram of wavelet decomposition of a fault of an outer ring of a rolling bearing provided by an embodiment of the application;
FIG. 6 is a wavelet decomposition five-layer signal diagram of a rolling bearing ball fault provided by the embodiment of the application;
FIG. 7 is a chart of a division of four operating ranges provided by an embodiment of the present application;
FIG. 8 is a graph of data amount and corresponding accuracy provided by an embodiment of the present application;
FIG. 9 is a graph of an index function provided in an embodiment of the present application;
FIG. 10 is a diagram illustrating the training results of an SVM support vector machine according to an embodiment of the present invention;
FIG. 11 is a diagram illustrating a test result of an SVM support vector machine provided by an embodiment of the present application;
fig. 12 is a schematic diagram of a rolling bearing fault diagnosis system based on a high-order origin moment according to an embodiment of the present application.
Detailed Description
The present invention will be described with reference to the accompanying drawings and embodiments.
At present, the rolling bearing is widely applied to rotary machinery, and the running state of the rolling bearing often directly influences the precision, the reliability and the service life of the whole machine. The service life of the rolling bearing is very discrete, and the rolling bearing cannot be maintained at regular time, so that the rolling bearing monitoring and fault diagnosis method has important significance for the state monitoring and fault diagnosis of the rolling bearing. At present, most of fault diagnosis and analysis of the rolling bearing are based on acquired signal attributes, however, most of the signals have the characteristics of nonlinearity, non-stability and the like, and information of the state of the rolling bearing can be obtained through analysis. Under the circumstance, how to utilize the information characteristics of the time domain signal, reduce the calculation links, simplify the fault diagnosis algorithm and improve the fault diagnosis efficiency is achieved, so that a method which can quickly and effectively extract the working condition fault characteristics and diagnose the fault is required to be found at present.
The principle and embodiment of a high-order central moment-based rolling bearing failure diagnosis method according to the present invention will be described in detail below. Referring to fig. 1, a schematic flow chart of a rolling bearing fault diagnosis method based on a high-order origin moment provided by an embodiment of the present application is shown, and referring to fig. 1, the method includes:
and S101, extracting sample data of normal working conditions and fault working conditions in the operation of the rolling bearing, and carrying out standardized processing on the data.
The bearing data used in the embodiment of the application is bearing data of the university of Keiss Caesand, the experimental bearing is an SKF bearing, and an acceleration vibration sensor is adopted to acquire vibration signal data. The sampling frequency was 12 khz. Single point failures are caused by electrical discharge machining. The fault size was 0.007 inches and the speed was 1750 rpm. The working conditions are divided into four types, namely normal working conditions, inner ring faults, outer ring faults and ball faults.
And extracting normal data samples and fault data samples of the equipment. Respectively standardizing sample data of normal working conditions and fault working conditions of the rolling bearing and drawing time domain waveforms as shown in figure 2; the normalization processing is mainly provided for data processing convenience, and the data are mapped into a range for processing, so that the data processing is more convenient and faster. The normalized calculation formula is as follows:
wherein x is*For the normalized data, x is the raw data, xmaxIs the maximum value, x, in the raw dataminIs the minimum value in the raw data.
And S102, dividing the time domain signal into five layers of signals by adopting a wavelet decomposition method.
In the embodiment of the application, db8 is selected as the wavelet basis of decomposition, and five layers of signals are decomposed, and the results are shown in fig. 3-6.
And S103, calculating the four-order origin moments of the normal working condition and the three fault working conditions.
The formula for calculating the fourth-order origin moment of the normal working condition and the three fault working conditions is specifically as follows:
wherein k is the fourth order origin moment, a is data, and n is the data size. Respectively calculating four-order origin moments of a normal working condition and a fault working condition, taking the data volume of 4000 as an example, and obtaining the following data:
TABLE 1 fourth moment of Normal working conditions
TABLE 2 fourth moment of inner circle failure
TABLE 3 fourth moment of ball failure
TABLE 4 outer lane failure mode fourth moment
S104, combining the four-order origin moments (1,2,3,4,5) of the five-layer signals obtained by calculation into two vectors, namely a vector A (1,2,3) and a vector B (3,4,5), by taking a vector included angle as a characteristic, and calculating an included angle between the vector A (1,2,3) and the vector B (3,4, 5).
The five-layer signals obtained by wavelet decomposition are calculated to obtain a fourth-order origin moment, and the five-layer signals obtain five fourth-order origin moments. Sequentially arranging to obtain a group of vectors (1,2,3,4,5), combining the vectors (1,2,3,4,5) into two vectors, namely a vector A (1,2,3) and a vector B (3,4,5), and calculating an included angle between the vector A (1,2,3) and the vector B (3,4, 5); the formula is as follows:
wherein theta represents an angle of rotation of the rotary shaft,
represents the vectors (1,2,3),
representing the vector (3,4, 5). Taking the data amount as 4000 as an example, the included angle between the vectors is calculated as follows:
TABLE 5 training data vector Angle
And S105, determining the distinguishing ranges of the four working conditions, and calculating the accuracy of different data volumes.
First, data in 10000 to 70000 is extracted as training data. The first step is to calculate the four-order origin moment of four working conditions; secondly, calculating a characteristic angle of each working condition; and calculating the average value of the training angles of each working condition to obtain four average angles of the four working conditions. The average angle of normal operation, the average angle of inner ring faults, the average angle of ball faults and the average angle of outer ring faults are sequentially.
After averaging again between two average angles of two adjacent working conditions, a distinguishing value can be calculated and used as the upper limit and the lower limit of a distinguishing range of the two adjacent working conditions; the average value between the average angle for normal operation and the average angle for inner ring failure may determine a difference value. The average value between the mean angle of inner ring failure and the mean angle of ball failure may determine a discrimination value. The average value between the mean angle of the ball failure and the mean angle of the outer ring failure can determine a discrimination value.
Then, determining the ranges of the four working conditions according to the distinguishing values, wherein the range of the normal working conditions is that theta is more than or equal to 54 and less than 100; the failure range of the inner ring is more than or equal to 19 and less than 54; the failure range of the ball is more than or equal to 9 and less than 19; the outer ring fault range is more than or equal to 0 and less than 9. The discrimination ranges are shown in fig. 7. After the range is determined, the angles calculated in the second step are respectively distinguished.
When the four working conditions are all in the corresponding ranges, the condition is judged to be correct. If one of the operating conditions is not within the corresponding range, it is determined that the condition is erroneous. The accuracy in each case of ten data amounts is shown in fig. 8.
S106, establishing an index function of the relation between the accuracy and the data volume, and selecting the optimal data volume.
The method specifically comprises the following steps:
wherein c is the correct rate, d is the data size, α is the adjustment coefficient, and y is the index function, wherein, when the index function image has a unique peak, the adjustment coefficient α is determined to be 0.6, and the peak is the function optimal point, and the corresponding data size is the determined optimal data size.
The adjustment coefficient α has a value range of (α ═ 0,0.05,0.1, …,1), d is the size of the data volume, and a value range of (d ═ 1000,2000, …, 10000).
Selecting data (10001-. The result of the index function of the relationship between the accuracy and the data amount is shown in fig. 9.
Table 6 test data vector angle
And S107, establishing an SVM (support vector machine) classifier, and inputting the training sample and the test sample into the SVM classifier respectively for classification diagnosis.
In the embodiment of the application, the identification result 1 represents a normal working condition; 2 denotes inner ring failure; 3 indicates ball failure; 4 represents an outer ring failure; according to the steps, the test samples are distinguished, namely, the fault diagnosis is realized. The results are shown in FIGS. 10 and 11.
Corresponding to the method for diagnosing the fault of the rolling bearing based on the high-order origin moment provided by the embodiment, the application also provides an embodiment of a system for diagnosing the fault of the rolling bearing based on the high-order origin moment.
Referring to fig. 12, the rolling bearing fault diagnosis system based on the high-order origin moment includes: an extraction module 201, a wavelet decomposition module 202, a first calculation module 203, a second calculation module 204, a determination module 205, a setup module 206, and a fault diagnosis module 207.
The extraction module 201 is configured to extract sample data of a normal working condition and a fault working condition in the operation of the rolling bearing, and perform standardized processing on the data. The wavelet decomposition module 202 is configured to divide the time domain signal into five layers of signals by using a wavelet decomposition method. The first calculating module 203 is used for calculating the fourth-order origin moments of the normal working condition and the three fault working conditions. The second calculating module 204 is configured to combine the calculated five-layer signals with the fourth-order origin moments (1,2,3,4,5) into two vectors, which are vector a (1,2,3) and vector B (3,4,5), and calculate an included angle between vector a (1,2,3) and vector B (3,4,5), by using a vector included angle as a feature. The determining module 205 is configured to determine the range of the four operating conditions, and calculate the accuracy of different data sizes. The establishing module 206 is configured to establish an index function of a relationship between the accuracy and the data size, and select an optimal data size. The fault diagnosis module 207 is used for establishing an SVM (support vector machine) classifier, and inputting the training samples and the test samples into the SVM classifier respectively for classification diagnosis
It is noted that, in this document, relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Of course, the above description is not limited to the above examples, and technical features that are not described in this application may be implemented by or using the prior art, and are not described herein again; the above embodiments and drawings are only for illustrating the technical solutions of the present application and not for limiting the present application, and the present application is only described in detail with reference to the preferred embodiments instead, it should be understood by those skilled in the art that changes, modifications, additions or substitutions within the spirit and scope of the present application may be made by those skilled in the art without departing from the spirit of the present application, and the scope of the claims of the present application should also be covered.
Claims (9)
1. A rolling bearing fault diagnosis method based on a high-order origin moment is characterized by comprising the following steps:
extracting sample data of normal working conditions and fault working conditions in the operation of the rolling bearing, and carrying out standardized processing on the data;
dividing the time domain signal into five layers of signals by adopting a wavelet decomposition method;
calculating four-order origin moments of a normal working condition and three fault working conditions;
combining the four-order origin moments (1,2,3,4,5) of the five-layer signals obtained by calculation into two vectors, namely a vector A (1,2,3) and a vector B (3,4,5), by taking a vector included angle as a characteristic, and calculating an included angle between the vector A (1,2,3) and the vector B (3,4, 5);
determining the distinguishing ranges of the four working conditions, and calculating the accuracy of different data volumes;
establishing an index function of the relation between the accuracy and the data volume, and selecting the optimal data volume;
and establishing an SVM (support vector machine) classifier, and respectively inputting the training sample and the test sample into the SVM classifier for classification diagnosis.
2. The rolling bearing fault diagnosis method based on the higher order origin moment according to claim 1, wherein the extracting of the sample data of the normal working condition and the fault working condition during the rolling bearing operation and the normalization of the data comprises:
wherein x is*The data is normalized; x is original data; x is the number ofmaxIs the maximum value in the original data; x is the number ofminIs the minimum value in the raw data.
3. The rolling bearing fault diagnosis method based on the higher-order origin moment as claimed in claim 1, wherein the dividing of the time domain signal into five layers of signals by the wavelet decomposition method comprises: a wavelet decomposition method is adopted to select the db8 wavelet in the time domain signals as a wavelet base, and the db8 wavelet is divided into five layers of signals.
4. The rolling bearing fault diagnosis method based on the higher-order origin moment according to claim 1, wherein the calculating of the fourth-order origin moment of the normal working condition and the three fault working conditions comprises:
wherein k is the fourth order origin moment, a is data, and n is the data size.
5. The rolling bearing fault diagnosis method based on the higher-order origin moment according to claim 1, wherein the vector included angle is used as a characteristic, the calculated five-layer signal fourth-order origin moment (1,2,3,4,5) is combined into two vectors, namely vector A (1,2,3) and vector B (3,4,5), and the included angle between vector A (1,2,3) and vector B (3,4,5) is calculated, and the method comprises the following steps:
the vector included angle is characterized in that five layers of signals obtained by wavelet decomposition are calculated to obtain a fourth order origin moment, five layers of signals obtain five fourth order origin moments, and the five layers of signals are perpendicular to each other in space, so that the fourth order origin moments calculated by the signals of each layer are combined into a group of space vectors;
sequentially arranging to obtain a group of vectors (1,2,3,4,5), combining the vectors (1,2,3,4,5) into two vectors, namely a vector A (1,2,3) and a vector B (3,4,5), and calculating an included angle between the vector A (1,2,3) and the vector B (3,4, 5); the formula is as follows:
wherein theta represents an angle of rotation of the rotary shaft,
represents the vectors (1,2,3),representing the vector (3,4, 5).
6. The rolling bearing fault diagnosis method based on the higher-order origin moment as claimed in claim 1, wherein the determining the distinguishing ranges of four working conditions and calculating the accuracy of different data size comprises:
calculating four-order origin moments of four working conditions;
calculating a characteristic angle of each working condition;
and taking the characteristic angle obtained by calculating the training data as a training characteristic, and then calculating the average value of the training angle of each working condition to obtain four average angles of the four working conditions, namely the average angle of normal operation, the average angle of the inner ring fault, the average angle of the ball fault and the average angle of the outer ring fault in sequence.
7. The method for diagnosing the fault of the rolling bearing based on the higher-order origin moment according to claim 1, wherein the establishing of the index function of the relationship between the accuracy and the data amount comprises:
where c is the accuracy, d is the data size, α is the adjustment factor, and y is the index function.
8. The rolling bearing fault diagnosis method based on the high-order origin moment as claimed in claim 1, wherein the establishing of the SVM classifier, the inputting of the training samples and the testing samples into the SVM classifier respectively for classification diagnosis, comprises:
establishing an SVM (support vector machine) classifier, and inputting the training sample and the test sample into the SVM classifier respectively for classification diagnosis;
and the identification result 1 represents a normal working condition, 2 represents an inner ring fault, 3 represents a ball fault, and 4 represents an outer ring fault, so that the fault diagnosis of the differentiation of the test samples is realized.
9. A rolling bearing fault diagnosis system based on a high-order origin moment, the system comprising:
the extraction module is used for extracting sample data of normal working conditions and fault working conditions in the operation of the rolling bearing and carrying out standardized processing on the data;
the wavelet decomposition module is used for dividing the time domain signals into five layers of signals by adopting a wavelet decomposition method;
the first calculation module is used for calculating the fourth-order origin moments of the normal working condition and the three fault working conditions;
the second calculation module is used for combining the calculated five-layer signals into two vectors, namely a vector A (1,2,3) and a vector B (3,4,5), by taking the included angle of the vectors as a characteristic, and calculating the included angle between the vector A (1,2,3) and the vector B (3,4, 5);
the determining module is used for determining the distinguishing ranges of the four working conditions and calculating the accuracy of different data volumes;
the establishing module is used for establishing an index function of the relation between the accuracy and the data volume and selecting the optimal data volume;
and the fault diagnosis module is used for establishing an SVM (support vector machine) classifier and inputting the training samples and the test samples into the SVM classifier respectively for classification diagnosis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911181169.0A CN110749443B (en) | 2019-11-27 | 2019-11-27 | Rolling bearing fault diagnosis method and system based on high-order origin moment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911181169.0A CN110749443B (en) | 2019-11-27 | 2019-11-27 | Rolling bearing fault diagnosis method and system based on high-order origin moment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110749443A true CN110749443A (en) | 2020-02-04 |
| CN110749443B CN110749443B (en) | 2021-05-18 |
Family
ID=69284703
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911181169.0A Expired - Fee Related CN110749443B (en) | 2019-11-27 | 2019-11-27 | Rolling bearing fault diagnosis method and system based on high-order origin moment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110749443B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111222495A (en) * | 2020-02-17 | 2020-06-02 | 济南大学 | Rolling bearing fault diagnosis method based on wavelet decomposition and improved Manhattan distance |
| CN112036458A (en) * | 2020-08-21 | 2020-12-04 | 上海电机学院 | Fault diagnosis method for rolling bearing |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080091361A1 (en) * | 2004-06-08 | 2008-04-17 | Schaeffler Kg | Method And Computer Programme For Determining Operating Parameters In A Roller Bearing And A Roller Bearing Which May Be Analysed Thus |
| CN103499445A (en) * | 2013-09-28 | 2014-01-08 | 长安大学 | Time-frequency slice analysis-based rolling bearing fault diagnosis method |
| CN104376504A (en) * | 2014-11-06 | 2015-02-25 | 国家电网公司 | Power distribution system probability reliability assessing method based on analytical method |
| CN105300693A (en) * | 2015-09-25 | 2016-02-03 | 东南大学 | Bearing fault diagnosis method based on transfer learning |
| CN107505133A (en) * | 2017-08-10 | 2017-12-22 | 滁州学院 | The probability intelligent diagnosing method of rolling bearing fault based on adaptive M RVM |
| CN108537260A (en) * | 2018-03-29 | 2018-09-14 | 上海交通大学 | A kind of crane transmission axis method for diagnosing faults and system |
| CN109708891A (en) * | 2019-01-30 | 2019-05-03 | 华南理工大学 | A kind of flexibility elliptic bearing raceway method for diagnosing faults |
-
2019
- 2019-11-27 CN CN201911181169.0A patent/CN110749443B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080091361A1 (en) * | 2004-06-08 | 2008-04-17 | Schaeffler Kg | Method And Computer Programme For Determining Operating Parameters In A Roller Bearing And A Roller Bearing Which May Be Analysed Thus |
| CN103499445A (en) * | 2013-09-28 | 2014-01-08 | 长安大学 | Time-frequency slice analysis-based rolling bearing fault diagnosis method |
| CN104376504A (en) * | 2014-11-06 | 2015-02-25 | 国家电网公司 | Power distribution system probability reliability assessing method based on analytical method |
| CN105300693A (en) * | 2015-09-25 | 2016-02-03 | 东南大学 | Bearing fault diagnosis method based on transfer learning |
| CN107505133A (en) * | 2017-08-10 | 2017-12-22 | 滁州学院 | The probability intelligent diagnosing method of rolling bearing fault based on adaptive M RVM |
| CN108537260A (en) * | 2018-03-29 | 2018-09-14 | 上海交通大学 | A kind of crane transmission axis method for diagnosing faults and system |
| CN109708891A (en) * | 2019-01-30 | 2019-05-03 | 华南理工大学 | A kind of flexibility elliptic bearing raceway method for diagnosing faults |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111222495A (en) * | 2020-02-17 | 2020-06-02 | 济南大学 | Rolling bearing fault diagnosis method based on wavelet decomposition and improved Manhattan distance |
| CN111222495B (en) * | 2020-02-17 | 2023-05-12 | 济南大学 | Rolling bearing fault diagnosis method based on wavelet decomposition and improved Manhattan distance |
| CN112036458A (en) * | 2020-08-21 | 2020-12-04 | 上海电机学院 | Fault diagnosis method for rolling bearing |
| CN112036458B (en) * | 2020-08-21 | 2023-05-23 | 上海电机学院 | Rolling bearing fault diagnosis method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110749443B (en) | 2021-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111353482B (en) | LSTM-based fatigue factor recessive anomaly detection and fault diagnosis method | |
| CN108444708B (en) | Method for establishing rolling bearing intelligent diagnosis model based on convolutional neural network | |
| CN112179691B (en) | Mechanical equipment running state abnormity detection system and method based on counterstudy strategy | |
| CN110749443B (en) | Rolling bearing fault diagnosis method and system based on high-order origin moment | |
| CN113834657A (en) | Bearing fault early warning and diagnosis method based on improved MSET and frequency spectrum characteristics | |
| CN116434372B (en) | Intelligent data acquisition system and working condition identification system for variable working condition equipment | |
| CN110441060A (en) | One kind being based on mechanical failure diagnostic method combined of multi-sensor information | |
| CN111122191B (en) | Equipment health alarm threshold setting method based on EWMA control | |
| Arabaci et al. | A knowledge-based diagnosis algorithm for broken rotor bar fault classification using FFT, principal component analysis and support vector machines | |
| CN111678699B (en) | Early fault monitoring and diagnosing method and system for rolling bearing | |
| CN110118657A (en) | Based on relative entropy and K nearest neighbor algorithm Fault Diagnosis of Roller Bearings and system | |
| KR101967065B1 (en) | Fault diagnosis apparatus and method for robust to environmental change | |
| CN108195584A (en) | A kind of Fault Diagnosis of Roller Bearings based on accuracy spectrogram | |
| CN107515118B (en) | Bearing in rotating machinery method for diagnosing faults | |
| CN111811819A (en) | Bearing fault diagnosis method and device based on machine learning | |
| KR20210006832A (en) | Method and apparatus for machine fault diagnosis | |
| CN111060302A (en) | Multi-parameter equipment state comprehensive monitoring and diagnosing device and method | |
| CN116226719A (en) | Bearing fault diagnosis method based on multidimensional steady-state vibration characteristics and related components | |
| CN110530631A (en) | A kind of gear list type fault detection method based on hybrid classifer | |
| Pazouki et al. | Fault diagnosis and condition monitoring of bearing using multisensory approach based fuzzy-logic clustering | |
| CN111222495B (en) | Rolling bearing fault diagnosis method based on wavelet decomposition and improved Manhattan distance | |
| CN114861749A (en) | Low-sample bearing fault diagnosis method based on depth prototype network | |
| CN114742093A (en) | Rolling bearing fault diagnosis method and device based on time-frequency curve extraction and classification | |
| TWI398629B (en) | Equipment supervision method | |
| CN113609964A (en) | Motor abnormal vibration early warning method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210518 Termination date: 20211127 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |