CN117076974A - Bearing fault parameter-free diagnosis method based on order spectrum analysis and classification model - Google Patents
Bearing fault parameter-free diagnosis method based on order spectrum analysis and classification model Download PDFInfo
- Publication number
- CN117076974A CN117076974A CN202310876592.2A CN202310876592A CN117076974A CN 117076974 A CN117076974 A CN 117076974A CN 202310876592 A CN202310876592 A CN 202310876592A CN 117076974 A CN117076974 A CN 117076974A
- Authority
- CN
- China
- Prior art keywords
- bearing
- energy
- order spectrum
- fault
- steps
- 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
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000003745 diagnosis Methods 0.000 title claims abstract description 38
- 238000013145 classification model Methods 0.000 title claims abstract description 33
- 238000010183 spectrum analysis Methods 0.000 title claims abstract description 16
- 238000001228 spectrum Methods 0.000 claims abstract description 87
- 238000012549 training Methods 0.000 claims abstract description 28
- 230000001133 acceleration Effects 0.000 claims abstract description 27
- 238000006243 chemical reaction Methods 0.000 claims abstract description 22
- 238000012545 processing Methods 0.000 claims abstract description 17
- 238000001914 filtration Methods 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000003595 spectral effect Effects 0.000 claims description 4
- 230000005856 abnormality Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 abstract description 5
- 238000005096 rolling process Methods 0.000 description 6
- 238000009826 distribution Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000002405 diagnostic procedure Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000002759 z-score normalization Methods 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention belongs to the technical field of signal processing, and particularly relates to a bearing fault parameter-free diagnosis method based on order spectrum analysis and a classification model, which comprises the following steps: the method comprises the steps of collecting actual vibration signals of a bearing, calculating the frequency conversion of the actual vibration signals through a speed frequency spectrum of the bearing, calculating a corresponding acceleration order spectrum according to the frequency conversion, and carrying out normalization processing on the order spectrum energy. And taking the orders and energy data of the bearings with different fault categories as characteristic parameters to input the model for classification training, so as to obtain a classification model. For the bearing signals of unknown faults, the order spectrum data of the bearing signals can be input into a classifier model to carry out classification diagnosis, and the corresponding fault types are finally obtained; the diagnosis method is driven by the vibration data of the bearing, can calculate the fault characteristic frequency and the frequency conversion information of the bearing, and constructs a corresponding classification model to diagnose the bearing fault; in addition, the method does not need prior parameters, has good applicability to different bearings, and can carry out accurate fault diagnosis on the bearings.
Description
Technical Field
The invention belongs to the technical field of signal processing, and particularly relates to a bearing fault parameter-free diagnosis method based on order spectrum analysis and a classification model.
Background
About 30% of the failures of the rotating machinery are caused by bearing failures, while about 90% of the failures of the rolling bearings are caused by failures of the outer ring and the inner ring, and if the bearing failures are not handled in time, the bearing failures have serious influence on the running of the equipment, so the significance of the diagnosis of the bearing failures is very important. First, bearing fault diagnosis can help us find bearing faults in time. The working state information of the bearing is obtained through equipment such as a vibration sensor and the like, so that whether the bearing has abnormal vibration, noise and other problems can be found, and whether the bearing has faults or not is judged. Second, bearing failure diagnosis can help us determine the cause of bearing failure. By analyzing the vibration signal, specific causes of bearing failure, such as wear, fatigue, corrosion, etc., can be determined. Finally, bearing failure diagnosis can help us formulate an effective maintenance solution. According to the reason and degree of bearing fault, corresponding maintenance schemes are formulated, including measures such as bearing replacement and abrasion part repair, so that normal operation of mechanical equipment is ensured.
In recent years, many researches on bearing fault diagnosis are performed, for example, a data-driven train bearing fault diagnosis method is proposed in a patent with publication number CN115855509a, and an acoustic signal of a bearing to be detected is obtained; determining a frequency domain cumulative value from the acoustic signal; based on a non-parametric probability regression model, determining a normalized logarithmic Bayesian factor according to the frequency domain cumulative value, and further determining a diagnosis result of the bearing to be detected; in the patent with publication number of CN114970044B, a rolling bearing fault diagnosis method and system based on a threshold convolutional neural network are provided, and rolling bearing vibration history data containing faults is trained to obtain a rolling bearing fault diagnosis model based on the threshold convolutional neural network; collecting vibration data of the rolling bearing in real time; monitoring vibration data in real time by using a trained fault diagnosis model, and identifying whether faults exist; if the fault exists, fault characteristics are extracted and classified by using a fault diagnosis model, and a fault identification result is output. In the patent publication No. CN114264477A, a method for diagnosing the defects of a rolling bearing is proposed, the central displacement of a ball when passing through a bearing raceway is solved, a bearing defect model is established, and the size of the defects is diagnosed according to vibration characteristics. The interaction of the balls passing through the defects is specifically classified, and then the subsequent vibration characteristic analysis is performed.
Existing methods or studies have one or more of the following drawbacks and disadvantages:
1. the bearing vibration signals contain various low-frequency and high-frequency noises, and if the noise signals cannot be effectively removed, the extraction of fault characteristic frequencies is influenced.
2. The machine learning method is directly adopted, so that whether the bearing has faults or not can only be determined, and the specific fault type can not be determined.
3. Different mechanical equipment's operating mode is different, and bearing vibration energy difference is great, and if direct classification training carries out, the error is great, and can't train out correct classification model.
4. The rotation speeds of different mechanical devices are different, so that the frequency of faults is different when the bearing performs spectrum analysis.
5. The parameters and fault characteristic coefficients of different bearings are different, the same type of parameter threshold is adopted to judge the fault type, and the fault type has a great error.
6. Although a method for calculating fault characteristic frequency according to a specific model of a bearing can obtain a relatively accurate bearing fault characteristic coefficient, in an actual application scene, the model of the bearing is more, bearing information is difficult to obtain in advance, and many diagnosis platforms cannot perform fault identification according to actual bearing parameters.
7. The fault type determination of the bearing depends on the accurate identification of the fault frequency and the frequency of rotation, and if the frequency of rotation and the fault frequency with harmonics cannot be effectively resolved, the specific fault type of the bearing cannot be accurately judged.
Disclosure of Invention
The invention aims to provide a bearing fault parameter-free diagnosis method based on an order spectrum analysis and classification model, so as to solve the problems in the background technology.
The invention realizes the above purpose through the following technical scheme:
a bearing fault parameter-free diagnosis method based on order spectrum analysis and classification model comprises the following steps:
s1, collecting and processing vibration signal data of a bearing to obtain a speed vibration signal and an acceleration vibration signal, sequentially carrying out Fourier transform and median filtering on the speed vibration signal to generate a speed frequency spectrum corresponding to the bearing, and extracting a rotating frequency;
s2, according to the acceleration vibration signal and the frequency conversion, obtaining acceleration order spectrum energy of a corresponding bearing, normalizing the order spectrum energy to generate training data, inputting the training data into a pre-constructed classifier for training, and obtaining a fault classification model;
s3, preprocessing bearing fault signal data to be identified, which are acquired in real time, so as to generate order spectrum energy corresponding to the bearing signals, inputting the order spectrum energy into the fault diagnosis model for diagnosis, and outputting final fault types.
As a further optimization scheme of the present invention, the specific steps in step S1 include:
s101, acquiring a speed vibration signal x [ n ] and an acceleration vibration signal a [ n ], and sequentially carrying out filtering analysis and abnormal data processing;
s102, carrying out Fourier transform on the speed vibration signal x [ n ] after the abnormality processing to obtain a complex frequency spectrum Px [ f ];
s103, median filtering is carried out on the complex spectrum Px [ f ] to obtain spectrum background energy Px_Base [ f ];
s104, calculating all peak points of the complex spectrum Pxf to obtain PxVel f;
s105, calculating the difference value between the peak point PxVel [ f ] and the background energy Px_Base [ f ] to obtain a difference value array Differ [ f ];
s106, selecting a position corresponding to the maximum value of the difference array Differ [ f ] to obtain the frequency conversion Frpm.
As a further optimization scheme of the present invention, the step S2 of calculating the order spectral energy comprises the following specific steps:
s201, performing fast Fourier transform on the acquired acceleration vibration signal a [ n ] to obtain a complex frequency spectrum PxA [ f ];
s202, dividing the abscissa of the complex spectrum PxA [ f ] by the frequency conversion to obtain a new coordinate axis, wherein the new coordinate axis corresponds to an x-axis [ x1, x2 ], which is the order spectrum energy.
As a further optimization scheme of the present invention, the step S2 of normalizing the order spectrum energy specifically includes the following steps:
s203, selecting the energy maximum E in the order spectrum energy max Energy minimum E min ；
S204, adopting the following formula to obtain energy E of each order i Calculating to obtain normalized energy y i ：
S205, repeating the steps S203 and S204, and finally obtaining the x-axis [ x1, x2, ], xi ] and the y-axis [ y1, y2, ], yi of the order spectrum energy.
As a further optimization scheme of the invention, existing bearing faults are classified: the method comprises the following specific steps of training a pre-constructed classifier in the step S2, wherein the specific steps comprise the following steps of:
s206, collecting bearing signal data containing the faults, and calculating corresponding order spectrum energy;
s207, inputting the order spectrum energy and the corresponding fault category into the classifier as a training set, and training a classification model S (x, y).
As a further optimization scheme of the present invention, the specific steps in step S3 include:
s301, collecting bearing vibration signals of unknown faults;
s302, calculating the order spectrum energy corresponding to the bearing signal according to the steps S1 and S2;
s303, inputting the order spectrum energy into a classification model S (x, y) obtained through training in the step S207 for diagnosis and classification, and obtaining a final class.
The invention has the beneficial effects that:
(1) According to the invention, through analyzing the acceleration and speed spectrums of the equipment to be tested, the frequency conversion and acceleration order spectrum energy of the equipment to be tested are obtained, and after normalization, a bearing fault classification model is trained, so that the non-parameter fault diagnosis classification of the bearing can be realized;
(2) According to the invention, the frequency conversion is analyzed and calculated in the velocity spectrum, no additional rotating speed data is needed, and the accurate frequency conversion can be obtained by analyzing the data of the device;
(3) According to the method, the order spectrum is analyzed in the acceleration spectrum, and the energy distribution of the order spectrum is normalized and adjusted, so that errors caused by the difference of the energy of the spectrum of equipment under different working conditions can be avoided;
(4) According to the invention, the acceleration order spectrum data is used as a training set of the classifier, so that the problem that the distribution difference of fault frequencies of bearing equipment with different rotating speeds in a frequency domain is large can be solved;
(5) According to the method provided by the invention, priori knowledge of the bearing is not needed, the rotating speed of equipment and the model of the bearing are not needed to be known, and fault diagnosis is carried out completely according to the data characteristics of the bearing data;
(6) The method for training the order spectrum data by adopting the classification model is firstly provided, the problem of different equipment characteristic frequency distribution of different rotating speeds can be effectively solved, and the classification accuracy can be effectively improved by adopting the classification model under the condition that the training data amount is enough, so that the specific type of bearing fault can be obtained.
Drawings
FIG. 1 is a schematic flow chart of the diagnostic method of the present invention;
FIG. 2 is a graph of velocity spectrum analysis within an actual processing routine in an embodiment of the present invention;
fig. 3 is a graph of acceleration orders within an actual processing routine in an embodiment of the present invention.
Detailed Description
The invention will now be described in further detail with reference to the accompanying drawings, wherein it is to be understood that the following detailed description is for the purpose of illustration only and is not to be construed as limiting the scope of the invention, as various insubstantial modifications and adaptations of the invention to those skilled in the art may be made in light of the foregoing disclosure.
Example 1
As shown in fig. 1, the present embodiment provides a method for diagnosing a bearing fault without parameters based on order spectrum analysis and classification model, which is driven by vibration data of a bearing, and can calculate fault characteristic frequency and frequency conversion information of the bearing, and construct a corresponding classification model to diagnose the bearing fault. In addition, the method does not need prior parameters, has good applicability to different bearings, and can carry out accurate fault diagnosis on the bearings.
The diagnosis method specifically comprises the following steps:
s1, collecting and processing vibration signal data of a bearing to obtain a speed vibration signal and an acceleration vibration signal, sequentially carrying out Fourier transform and median filtering on the speed vibration signal to generate a speed frequency spectrum corresponding to the bearing, and extracting a rotating frequency;
further, the specific steps in step S1 include:
s101, acquiring a speed vibration signal x [ n ] and an acceleration vibration signal a [ n ], and sequentially carrying out filtering analysis and abnormal data processing;
s102, carrying out Fourier transform on the speed vibration signal x [ n ] after the abnormality processing to obtain a complex frequency spectrum Px [ f ];
s103, median filtering is carried out on the complex spectrum Px [ f ] to obtain spectrum background energy Px_Base [ f ];
s104, calculating all peak points of the complex spectrum Pxf to obtain PxVel f;
s105, calculating the difference value between the peak point PxVel [ f ] and the background energy Px_Base [ f ] to obtain a difference value array Differ [ f ];
s106, selecting a position corresponding to the maximum value of the difference array Differ [ f ] to obtain the frequency conversion Frpm.
In the embodiment, the frequency conversion is analyzed and calculated in the velocity spectrum, no additional rotational speed data is needed, and the accurate frequency conversion can be obtained by analyzing the data of the frequency conversion.
S2, according to the acceleration vibration signal and the frequency conversion, obtaining acceleration order spectrum energy of a corresponding bearing, normalizing the order spectrum energy to generate training data, inputting the training data into a pre-constructed classifier for training, and obtaining a fault classification model;
in this embodiment, the adoption of the order spectrum analysis can avoid the influence of the rotational speed appearing in the regression model or the envelope spectrum adopted in the prior art on the fault frequency.
Further, in step S2, the step of calculating the order spectral energy includes the following steps:
s201, performing fast Fourier transform on the acquired acceleration vibration signal a [ n ] to obtain a complex frequency spectrum PxA [ f ];
s202, dividing the abscissa of the complex spectrum PxA [ f ] by the frequency conversion to obtain a new coordinate axis, wherein the new coordinate axis corresponds to an x-axis [ x1, x2 ], which is the order spectrum energy.
In step S2, the step of normalizing the order spectrum energy specifically includes the following steps:
s203, selecting the energy maximum E in the order spectrum energy max Energy minimum E min ；
S204, adopting the following formula to obtain energy E of each order i Calculating to obtain normalized energy yi:
s205, repeating the steps S203 and S204, and finally obtaining the x-axis [ x1, x2, ], xi ] and the y-axis [ y1, y2, ], yi of the order spectrum energy.
Further, existing bearing faults are classified, and different faults are defined correspondingly: the method comprises the following specific steps of training a pre-constructed classifier in the step S2, wherein the specific steps comprise the following steps of:
s206, collecting bearing signal data containing the faults, and calculating the order spectrum energy of the corresponding faults;
the bearing signal data corresponding to the faults are bearing data with the type 4 faults selected from actual industrial field data and bearing data sets of the university of West storage.
S207, inputting the order spectrum energy and the corresponding fault category into the classifier as a training set, and training a classification model S (x, y).
In the embodiment, the order spectrum is analyzed in the acceleration spectrum, and the energy distribution of the order spectrum is normalized and adjusted, so that errors caused by the energy difference of equipment spectrums under different working conditions can be avoided; in addition, the acceleration order spectrum data is used as a training set of the classifier, so that the problem that the fault frequency of bearing equipment with different rotating speeds is large in distribution difference in a frequency domain can be solved.
S3, preprocessing bearing fault signal data to be identified, which are acquired in real time, so as to generate order spectrum energy corresponding to the bearing signals, inputting the order spectrum energy into the fault diagnosis model for diagnosis, and outputting final fault types.
In this embodiment, preprocessing is specifically performed on the bearing fault signal data to be identified, and the steps S1 and S2 are repeated to generate the order spectrum energy.
Further, the specific steps in step S3 include:
s301, collecting bearing vibration signals of unknown faults;
s302, calculating the order spectrum energy corresponding to the bearing signal according to the steps S1 and S2;
s303, inputting the order spectrum energy into a classification model S (x, y) obtained through training in the step S207 for diagnosis and classification, and obtaining a final class.
It can be understood that the diagnosis method provided by the embodiment does not need prior knowledge of the bearing, does not need to know the rotating speed of equipment and the model of the bearing, and completely performs fault diagnosis according to the data characteristics of the bearing data.
The above method is further elaborated in connection with the actual processing routines.
Case:
1. the sensor collects vibration data of the actual bearing on site (the vibration data can be data which is collected on site in real time and has known fault types) and inputs the vibration data to the server;
2. calculating a speed spectrum corresponding to the vibration data, and extracting a conversion frequency Frpm, wherein the speed spectrum is shown in figure 2;
as can be analyzed from fig. 2, the maximum value of the difference array diff f occurs at 20Hz, i.e. the machine has a 20Hz frequency of rotation;
3. and (3) calculating a complex frequency spectrum PxA [ f ] of the acceleration signal, and converting the complex frequency spectrum into a corresponding acceleration order spectrum according to the frequency conversion in the step (2). As shown in particular in figure 3 below.
4. Normalizing the order energy according to step S2;
5. the steps are adopted, bearing order spectrum data of different fault types are analyzed, and the bearing order spectrum data and fault types are input into the classification model S (x, y) together as a training set.
6. And (3) for the bearing signal data of the unknown faults obtained through collection, corresponding order spectrum data are calculated, and corresponding fault categories can be identified through the classification model S (x, y).
According to the processing results of the cases, the method obtains the frequency conversion and the acceleration order spectrum energy through analysis of the acceleration and the speed spectrum of the bearing, normalizes and trains a bearing fault classification model, and can realize the non-parameter fault diagnosis and classification of the bearing. The main parameters and the model of the method are obtained by data self-driven calculation, and the method can accurately diagnose the faults of the bearings of unknown signals.
It should be noted that the type of data processed in the diagnostic method of the present invention is not limited, and may be bearing vibration data, gear vibration data, or similar index data having a fixed fault characteristic.
The normalization method and the classification model proposed in step S2 are not limited in this embodiment, and may be normalization methods such as 0-1 normalization and z-score normalization; an class model such as SVM, KNN, LDA is also possible.
The algorithm mentioned in this embodiment may be implemented in the upper computer software in a signal processing manner, or may be implemented in other manners, such as a digital chip, a hardware circuit, etc.
The above examples merely represent a few embodiments of the present invention, which are described in more detail and are not to be construed as limiting the scope of the present invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention.
Claims (6)
1. A bearing fault parameter-free diagnosis method based on order spectrum analysis and classification model is characterized by comprising the following steps:
s1, collecting and processing vibration signal data of a bearing to obtain a speed vibration signal and an acceleration vibration signal, sequentially carrying out Fourier transform and median filtering on the speed vibration signal to generate a speed frequency spectrum corresponding to the bearing, and extracting a rotating frequency;
s2, according to the acceleration vibration signal and the frequency conversion, obtaining acceleration order spectrum energy of a corresponding bearing, normalizing the order spectrum energy to generate training data, inputting the training data into a pre-constructed classifier for training, and obtaining a fault classification model;
s3, preprocessing bearing fault signal data to be identified, which are acquired in real time, so as to generate order spectrum energy corresponding to the bearing signals, inputting the order spectrum energy into the fault diagnosis model for diagnosis, and outputting final fault types.
2. The method for diagnosing bearing faults without parameters based on the order spectrum analysis and classification model as claimed in claim 1, wherein the method is characterized by comprising the following steps of: the specific steps in the step S1 include:
s101, acquiring a speed vibration signal x [ n ] and an acceleration vibration signal a [ n ], and sequentially carrying out filtering analysis and abnormal data processing;
s102, carrying out Fourier transform on the speed vibration signal x [ n ] after the abnormality processing to obtain a complex frequency spectrum Px [ f ];
s103, median filtering is carried out on the complex spectrum Px [ f ] to obtain spectrum background energy Px_Base [ f ];
s104, calculating all peak points of the complex spectrum Pxf to obtain PxVel f;
s105, calculating the difference value between the peak point PxVel [ f ] and the background energy Px_Base [ f ] to obtain a difference value array Differ [ f ];
s106, selecting a position corresponding to the maximum value of the difference array Differ [ f ] to obtain the frequency conversion Frpm.
3. The method for diagnosing bearing faults without parameters based on the order spectrum analysis and classification model as claimed in claim 2, wherein the method is characterized by comprising the following steps of: in step S2, the order spectral energy is calculated, and the specific steps include the following:
s201, performing fast Fourier transform on the acquired acceleration vibration signal a [ n ] to obtain a complex frequency spectrum PxA [ f ];
s202, dividing the abscissa of the complex spectrum PxA [ f ] by the frequency conversion to obtain a new coordinate axis, wherein the new coordinate axis corresponds to an x-axis [ x1, x2 ], which is the order spectrum energy.
4. A method for diagnosing bearing failure without parameters based on an order spectrum analysis and classification model according to claim 3, wherein the method comprises the following steps: in step S2, the step of normalizing the order spectrum energy specifically includes the following steps:
s203, selecting the orderEnergy maximum E in spectral energy max Energy minimum E min ；
S204, adopting the following formula to obtain energy E of each order i Calculating to obtain normalized energy y i ：
S205, repeating the steps S203 and S204, and finally obtaining the x-axis [ x1, x2, ], xi ] and the y-axis [ y1, y2, ], yi of the order spectrum energy.
5. The method for diagnosing bearing faults without parameters based on the order spectrum analysis and classification model as claimed in claim 4, wherein the method is characterized by comprising the following steps of: classifying the existing bearing faults: the method comprises the following specific steps of training a pre-constructed classifier in the step S2, wherein the specific steps comprise the following steps of:
s206, collecting bearing signal data containing the faults, and calculating corresponding order spectrum energy;
s207, inputting the order spectrum energy and the corresponding fault category into the classifier as a training set, and training a classification model S (x, y).
6. The method for diagnosing bearing faults without parameters based on the order spectrum analysis and classification model as claimed in claim 5, wherein the method is characterized by comprising the following steps of: the specific steps in the step S3 include:
s301, collecting bearing vibration signals of unknown faults;
s302, calculating the order spectrum energy corresponding to the bearing signal according to the steps S1 and S2;
s303, inputting the order spectrum energy into a classification model S (x, y) obtained through training in the step S207 for diagnosis and classification, and obtaining a final class.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310876592.2A CN117076974A (en) | 2023-07-18 | 2023-07-18 | Bearing fault parameter-free diagnosis method based on order spectrum analysis and classification model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310876592.2A CN117076974A (en) | 2023-07-18 | 2023-07-18 | Bearing fault parameter-free diagnosis method based on order spectrum analysis and classification model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117076974A true CN117076974A (en) | 2023-11-17 |
Family
ID=88705144
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310876592.2A Pending CN117076974A (en) | 2023-07-18 | 2023-07-18 | Bearing fault parameter-free diagnosis method based on order spectrum analysis and classification model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117076974A (en) |
-
2023
- 2023-07-18 CN CN202310876592.2A patent/CN117076974A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Huo et al. | Incipient fault diagnosis of roller bearing using optimized wavelet transform based multi-speed vibration signatures | |
| CN110276416B (en) | Rolling bearing fault prediction method | |
| Wang et al. | A novel statistical time-frequency analysis for rotating machine condition monitoring | |
| CN113834657B (en) | Bearing fault early warning and diagnosis method based on improved MSET and frequency spectrum characteristics | |
| US6802221B2 (en) | System and method for conditioned-based monitoring of a bearing assembly | |
| US10520397B2 (en) | Methods and apparatuses for defect diagnosis in a mechanical system | |
| CN111365251B (en) | Intelligent diagnosis method for faults of centrifugal pump unit | |
| CN113092113B (en) | Rolling bearing fault diagnosis system based on spectral value ratio | |
| CN110398384A (en) | The machine health monitoring of rotating machinery | |
| Lu et al. | CEEMD-assisted bearing degradation assessment using tight clustering | |
| CN110196165B (en) | Rotary machine ball bearing fault diagnosis method based on K-means clustering and evidence reasoning | |
| CN116089813A (en) | Wind turbine generator set transmission chain fault diagnosis method based on machine learning | |
| CN117786461A (en) | Water pump fault diagnosis method, control device and storage medium thereof | |
| CN112380782A (en) | Rotating equipment fault prediction method based on mixed indexes and neural network | |
| CN110749443B (en) | Rolling bearing fault diagnosis method and system based on high-order origin moment | |
| Martins et al. | Improved variational mode decomposition for combined imbalance-and-misalignment fault recognition and severity quantification | |
| CN111456915A (en) | Fault diagnosis device and method for internal components of fan engine room | |
| CN117093938A (en) | Fan bearing fault detection method and system based on deep learning | |
| CN117076974A (en) | Bearing fault parameter-free diagnosis method based on order spectrum analysis and classification model | |
| Madhukar et al. | A Machine Learning Based Methodology for Fault Diagnosis in Rotating Machine | |
| CN114184375A (en) | Intelligent diagnosis method for common faults of gear box | |
| Nouioua et al. | Self-Organizing Maps and VMD for Accurate Diagnosis of Bearing Defects | |
| Wu et al. | Automatic Diagnosis of Rolling Element Bearing Under Different Conditions Based on RVMD and Envelope Order Capture | |
| TWI398629B (en) | Equipment supervision method | |
| Astolfi et al. | Wind turbine drive-train condition monitoring through tower vibrations measurement and processing |
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 |