CN109655266B - Wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis - Google Patents
Wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis Download PDFInfo
- Publication number
- CN109655266B CN109655266B CN201811507395.9A CN201811507395A CN109655266B CN 109655266 B CN109655266 B CN 109655266B CN 201811507395 A CN201811507395 A CN 201811507395A CN 109655266 B CN109655266 B CN 109655266B
- Authority
- CN
- China
- Prior art keywords
- fault
- spectrum
- avmd
- pseudo
- signal
- 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.)
- Active
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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis, and provides a novel large wind turbine generator bearing fault diagnosis method aiming at the problems of inaccurate decomposition effect and low calculation efficiency of the conventional bearing fault diagnosis method at the present stage. The core idea of the algorithm is as follows: firstly, decomposing different types of fault signals through an improved VMD algorithm, namely an AVMD algorithm, then using a PCA dimension reduction and denoising decomposition mode, carrying out spectrum transformation on principal components processed by the PCA, converting the principal components into a frequency domain, connecting the obtained frequency spectrums of the modes end to obtain a spectrum vector of fault characteristics, and constructing a fault characteristic library. And then carrying out the same processing on the signals to be detected. Fault diagnosis is accomplished using spectral correlation analysis. Compared with the traditional wind turbine generator rolling bearing fault diagnosis method, the method is more accurate, the calculation speed is higher, and the practical value is better.
Description
Technical Field
The invention relates to a fault diagnosis method applied to a rolling bearing of a wind turbine generator or a rolling bearing of large-scale mechanical equipment, which is particularly used for enhancing the signal decomposition accuracy aiming at the characteristics of non-stability and non-linearity of a vibration signal; belonging to the technical field of fault diagnosis based on data driving.
Background
With the increasing requirements of human beings on energy, the rapid development of the power industry, the wind power generation industry has become the main development trend of clean energy with the advantages of relatively low cost, abundant wind power resources, green and environment-friendly energy and the like. The maintenance cost after the wind power plant is built directly determines the benefit of the wind power plant, the wind power generation project is a project with long investment time, about 7 years is probably, and the profit period is long and even exceeds 10 years. Wind generating sets operating for long periods of time require regular maintenance and repair to ensure operational stability and safety. When the service life of the wind generating set in the project is 20 years, the maintenance cost of the wind generating set accounts for 10% -15% of the overall income; the operation and maintenance cost required for installing the wind generating set on the sea accounts for 20-25% of the overall benefit, and a large amount of operation and maintenance cost increases the operation cost of the project and reduces the economic benefit of the project. To maximize the efficiency of wind farms, it is desirable to minimize the operational and maintenance costs. The rolling bearing is one of the vital parts of the wind driven generator and one of the vital failure sources in the wind turbine gearbox. Statistics have shown that about 30% of mechanical failures are caused by rolling bearings, and 20% of motor failures are caused by rolling bearings. In addition, most wind generating sets in practical engineering are installed in regions with sufficient wind resources, such as grassland, gobi desert, desert and other environments, the range of installation of the wind generating sets is wide, the number of the wind generating sets is large, the wind generating sets is influenced by severe natural environments, and therefore the rolling bearings are prone to failure. Once the fan trouble is not handled in time, then cause the loss of electric power energy slightly, then cause the condemning of mechanical equipment and casualties seriously. Therefore, the method has great significance for timely completing fault diagnosis on the rolling bearing of the wind turbine generator.
The method is a good method for analyzing the vibration signals near the rolling bearing of the fan so as to complete fault diagnosis. But the vibration signal of the sensor often has non-steady and non-linear characteristics, so that the fault information in the signal is difficult to sufficiently mine. It is crucial to find a suitable method of signal analysis. Wavelet analysis, EMD, EEMD, etc. have been proposed as methods for analyzing signals. And signal characteristics are obtained through signal time-frequency analysis of the front end, and fault diagnosis is completed by combining a proper rear-end mode identification method. VMD is a new proposed signal time-frequency analysis method, which is superior to wavelet analysis in terms of decomposition of different frequency components, EMD, EEMD methods are spectrum comparison diagrams of several signal processing methods as shown in fig. 1 to 4. However, the number N of decomposition modes and the penalty parameter epsilon of the conventional VMD algorithm need to be set artificially, and if the two parameters are not well selected, the decomposition effect is greatly influenced. In response to this problem, it is necessary to improve the conventional VMD and then perform fault diagnosis by combining a suitable pattern recognition method.
Disclosure of Invention
The invention provides a novel large-scale wind turbine generator bearing fault diagnosis method aiming at the problems of inaccurate decomposition effect and low calculation efficiency of the conventional bearing fault diagnosis method at the present stage. The core idea of the algorithm is as follows: firstly, decomposing different types of fault signals through an improved VMD algorithm, namely an AVMD algorithm, then using a PCA dimension reduction and denoising decomposition mode, carrying out spectrum transformation on principal components processed by the PCA, converting the principal components into a frequency domain, connecting the obtained frequency spectrums of the modes end to obtain a spectrum vector of fault characteristics, and constructing a fault characteristic library. And then carrying out the same processing on the signals to be detected. Fault diagnosis is accomplished using spectral correlation analysis. Compared with the traditional wind turbine generator rolling bearing fault diagnosis method, the method is more accurate, the calculation speed is higher, and the practical value is better.
The invention adopts the technical scheme that the wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis is realized by the following steps:
A. establishing a fault library stage:
step 1) data acquisition and division: the method comprises the steps of collecting original vibration signals of M types of faults, dividing kth type of fault signals, wherein k is 1,2, … and M, dividing each type of fault signals into H sections, and enabling the lengths of the sections to be consistent. Selecting a front S section as a training sample of two diagnostic algorithms for training, selecting a rear K section as a detection sample to verify the effectiveness of the algorithms, wherein S + K is more than 0 and less than or equal to H;
step 2) parameter selection: firstly, carrying out short-time Fourier transform on an original signal, drawing a short-time Fourier spectrogram, and obtaining the number N of modes according to an image result. And then, gradually changing the size of the penalty parameter epsilon from 5 to 50 by adopting the correction step length tau to decompose the original signal, and carrying out correlation analysis on the superposed frequency spectrum of the decomposition result and the original frequency spectrum, wherein the penalty parameter epsilon corresponding to the highest correlation coefficient is the result.
Step 3), extracting fault pseudo-frequency spectrum characteristics: firstly, resolving M-type fault signals by using the AVMD, and resolving each fault signal to obtain N modal components. And (3) processing the decomposition mode of the AVMD by using PCA to obtain N principal components, selecting the first N principal components with the highest contribution rate, wherein N is less than N, and achieving the purposes of reducing dimensions and denoising. And performing fast Fourier transform on the n principal components, and then sequentially connecting the obtained frequency spectrums end to obtain a pseudo-frequency spectrum characteristic vector, wherein the pseudo-frequency spectrum characteristic vector is used for cross-correlation analysis.
B. The diagnosis phase is carried out:
and obtaining a fault pseudo-frequency spectrum characteristic vector for the unknown fault signal, and respectively solving cross correlation coefficients of the pseudo-frequency spectrum characteristic vector and all fault pseudo-frequency spectrum characteristic vectors in a fault characteristic library, wherein the fault type corresponding to the maximum cross correlation coefficient is a diagnosis result. The cross-correlation coefficient r is calculated as follows:
wherein X represents a pseudo-spectrum feature vector of an unknown fault signal, and Y represents a pseudo-spectrum feature vector of a fault type in the fault feature set. Cov represents the covariance of the two pseudo-spectral feature vectors. σ represents the standard deviation. The characteristic frequencies of the same fault have strong correlation. The correlation coefficient ranges from-1 to 1, with values closer to 1 indicating greater correlation and values closer to 0 indicating less correlation, or even no correlation.
Compared with the prior art, the invention provides a new method for diagnosing the fault of the rolling bearing of the fan by AVMD-spectrum correlation analysis. The method overcomes the defect that the common VMD algorithm manually selects two parameters of the modal number and the penalty parameter. And the purpose of reducing dimensions, removing noise and accurately selecting fault characteristics is achieved by using the PCA to process the AVMD decomposition result. Finally, spectrum correlation analysis is used for comparing the fault characteristic spectrum model base with the spectrum characteristics of the signal to be detected, the fault type corresponding to the maximum correlation coefficient is screened, fault diagnosis is completed, the operation speed is high, and the diagnosis efficiency is improved.
Drawings
Fig. 1 is a diagram of a VMD decomposition signal spectrum.
Fig. 2 is a diagram of an EMD decomposed signal spectrum.
FIG. 3 is a graph of EEMD decomposed signal spectra.
Fig. 4 is a diagram of a wavelet decomposition signal spectrum.
Fig. 5 is a failure experiment platform.
Fig. 6 is a flowchart showing the algorithm.
Fig. 7 is a short-time fourier spectrogram of a selected number of modes.
Figure 8 is a diagram of the optimal penalty parameter selection.
Fig. 9 is a graph comparing the superimposed spectrum of the time component with the original signal spectrum with the penalty parameter epsilon equal to 100.
Fig. 10 is a graph comparing the superimposed spectrum of the time component with the original signal spectrum with the penalty parameter epsilon of 2000.
Fig. 11 is a comparison graph of the superimposed spectrum of time components with the original signal spectrum with the penalty parameter epsilon being 5000.
Fig. 12 is a graph of the contribution ratio of each principal component.
FIG. 13 is a graph comparing the outer and inner ring characteristic spectra after PCA processing.
Fig. 14 is a 0.007 outer ring fault diagnosis result graph.
Fig. 15 is a 0.007 inner ring failure diagnosis result graph.
Fig. 16 is a 0.007 ball failure diagnosis result graph.
FIG. 17 is a diagram of 0.021 outer ring fault diagnosis result.
Fig. 18 is a 0.021 inner ring fault diagnosis result diagram.
FIG. 19 is a diagram of the 0.021 ball failure diagnosis result.
Fig. 20 is a diagram of the results of the wind field data outer ring fault diagnosis.
Fig. 21 is a diagram of a wind field data inner ring fault diagnosis result.
Detailed Description
The invention aims to mainly solve the problems of insufficient accuracy of diagnosis results, large calculated amount and low diagnosis efficiency of the traditional rolling bearing fault diagnosis method.
A wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis is characterized by comprising two stages of 'fault library modeling' and 'implementation diagnosis', and specifically comprises the following steps:
A. establishing a fault library stage:
1) a fault signal decomposition stage;
and decomposing the fault signal into N modal components by using an adaptive variational modal decomposition method (AVMD), and repeatedly decomposing each fault signal by using the AVMD for M types of fault signals for M times. Obtaining M groups of signal component sets with the number of N;
2) a characteristic spectrum acquisition stage;
firstly, carrying out Principal Component Analysis (PCA) denoising and dimensionality reduction processing on modal components of M fault signals, selecting the first n principal component components with the highest contribution rate, and converting the principal component components into frequency spectrums by using Fast Fourier Transform (FFT). And connecting the frequency spectrums of the n principal component components end to form a fault characteristic pseudo frequency spectrum. For M types of fault signals, a fault characteristic pseudo-spectrum vector needs to be constructed for each type of fault signal. The set a containing these M feature pseudo-spectrum vectors is called the fault feature set.
B. The diagnosis phase is carried out:
for unknown fault signals, decomposing the unknown fault signals into N intrinsic mode components by using an AVMD method, reducing and de-noising the N intrinsic mode components into N principal component components by using PCA, changing the first N principal component components with the highest contribution rate into frequency spectrums by using fast Fourier transform, and forming pseudo-spectrum feature vectors by connecting the N frequency spectrums end to end by N1. And (4) respectively solving the cross correlation coefficient of the pseudo-spectrum feature vector and all fault pseudo-spectrum feature vectors in the fault feature set A, wherein the fault type corresponding to the maximum cross correlation coefficient is the diagnosis result. The cross-correlation coefficient r is calculated as follows:
wherein X represents a pseudo-spectrum feature vector of an unknown fault signal, and Y represents a pseudo-spectrum feature vector of one fault type in the fault feature set. Cov represents the covariance of the two pseudo-spectral feature vectors. σ represents the standard deviation.
Determining the number of the modes. Before decomposing the signal variation mode, drawing a short-time Fourier transform spectrogram by using short-time Fourier transform, and acquiring the mode number N according to the image.
Determination of penalty parameter epsilon. And changing penalty parameters in the variation modal decomposition by adopting the correction step length tau, combining modal results after each decomposition, and solving a cross-correlation coefficient with an original signal. And obtaining an optimal penalty parameter epsilon according to the inflection point of the cross-correlation coefficient curve.
And thirdly, decomposing the fault signal by using the AVMD. And setting AVMD parameters according to the determined number N of the eigenmodes and the optimal penalty parameter epsilon, and then decomposing the fault signal by using the AVMD to obtain component signals of different modes.
Examples
The invention uses two data successively to prove the effectiveness of the algorithm. The first type of data uses a laboratory fan transmission chain platform to collect fault data and normal operation data of a bearing outer ring, a bearing inner ring and a bearing ball; the second type of data collects bearing inner ring, bearing outer ring and normal operation data of the real wind turbine generator of the wind power plant.
The following are two data related introductions:
the laboratory platform bearing data experimental data is that a single point fault is processed on a bearing by an electric spark technology, the type of the bearing is SKF6205, and an acceleration sensor is used for measuring a bearing vibration signal. The data comprises a plurality of groups of data under different conditions, and the vibration signals of the bearing driving end with the load of 3HP, the rotating speed of 1730rpm and the sampling frequency of 12000Hz are selected for algorithm verification. The running states of the bearing corresponding to the used data comprise four types of normal, inner ring fault, outer ring fault and rolling body fault, and the damage diameters comprise 0.007 inches and 0.021 inches. FIG. 5 is a failure testing platform.
The bearing fault data of the wind driven generator collected by the inner Mongolia grey-rising beam wind power plant (all fan models are Yangming 1.5MW fans) are divided into three types of data including outer ring faults, inner ring faults and normal signals, the sampling frequency is 26kHz, and the bearing model is 6332MC3SKF deep groove ball bearing. The rolling bearing specific parameters are shown in table 1.
TABLE 1 Rolling bearing 6332MC3SKF basic parameters
The method of the invention is used for realizing fault diagnosis of the rolling bearing, and mainly comprises two major steps of establishing a fault library and implementing diagnosis, for example, fig. 6 is a specific flow chart of the invention, and the following is specifically stated:
A. establishing a fault library stage:
step 1: for each fault signal of the experimental platform data, 12000 points in the raw signal data for each fault were divided into 300 samples, each sample containing 400 points. For each fault signal of the wind farm data, 12000 points in the raw signal data for each fault were divided into 300 samples, each sample containing 400 points.
Step 2: in order to determine the number of modes of the fault signal, the fault signal is converted into a short-time fourier spectrogram by using a short-time fourier transform function (tfrstft), and the number of decomposition modes N is determined to be 4 from the short-time fourier spectrogram by taking 0.007 inner ring fault signal as an example, and the mode selection process is shown in fig. 7. Looking at the frequency vertical axis direction of the short-time fourier spectrogram, it can be seen that the signal contains 4 frequency components.
And step 3: the initial value epsilon of the penalty parameter is set to 5, and the correction step tau is set to 50. The initial value ii of the number of cycles is set to 1, and the total number of cycles nn is set to 100. And preliminarily decomposing the fault signal by using the VMD, performing fast Fourier transform on the decomposed components respectively, and superposing the obtained component frequency spectrums. Then, the component superposition frequency spectrum and the original signal frequency spectrum are subjected to correlation analysis to obtain a correlation coefficient r1. And ii +1 enters the next circulation process until ii is nn. And finally screening a penalty factor corresponding to the maximum correlation coefficient according to the inflection point of the correlation coefficient graph. The correlation coefficient calculation formula is as follows:
where x (t) and y (t) represent the original signal spectrum and the component superposition spectrum, respectively, and σ represents the standard deviation.
Taking 0.007 inner ring fault signal as an example, the penalty parameter selection process of AVMD is shown in fig. 8, and in the process of converting the penalty parameter from 0 to 4000, it can be seen that when the penalty factor epsilon is 2000, the spectral correlation between the decomposed component superposition spectrum and the original signal is the highest. Referring to fig. 9, fig. 10, and fig. 11, comparing the 0.007 inner ring fault component superposition spectrum with the original signal spectrum when the penalty parameters 100, 2000 and 5000 are respectively selected, it can be seen that the coincidence degree of the two spectra is the highest when the penalty parameter is 2000, indicating that the decomposition effect is the most accurate when the penalty parameter is 2000.
And 4, step 4: processing modal components of the M-type fault signal decomposition by using a PCA algorithm to obtain 4 principal components, wherein FIG. 12 shows the contribution rate of each principal component, selecting the first 3 principal components with the highest contribution rate to perform fast Fourier transform on the principal components and converting the principal components into frequency spectrums, and connecting the frequency spectrums end to end according to the contribution rate from high to low to form pseudo-spectrum feature vectors. As shown in fig. 13, which is a comparison graph of the inner circle pseudo-spectral feature vector and the outer circle pseudo-spectral feature vector, it can be seen that the different feature vectors have a distinct difference. Forming a set A by the M types of pseudo-spectrum feature vectors, wherein the A is a spectrum feature library;
B. and (3) implementing a fault diagnosis stage:
and (3) carrying out the same processing of the steps (1) to (4) on the unknown fault signal to obtain a pseudo-spectrum characteristic vector to be detected, and carrying out cross-correlation analysis on the pseudo-spectrum characteristic vector to be detected and the pseudo-spectrum characteristic vectors of M types of faults obtained in the step (4) respectively to obtain M cross-correlation coefficients r, wherein the fault type corresponding to the maximum r value is the final diagnosis result.
The steps are the specific application of the method in the fault of the rolling bearing. In order to verify the effectiveness of the method, fault diagnosis experiments are carried out on fault data of 0.007 outer ring, 0.007 inner ring, 0.007 ball, 0.021 outer ring, 0.021 inner ring and 0.021 ball in test bed data, and then fault diagnosis experiments are carried out again by using the inner ring data and the outer ring data in the wind field data. The experimental results obtained using the experimental platform data are shown in fig. 14 to 19, and the diagnostic results obtained using the real fan data are shown in fig. 20 to 21. Each graph includes a probability magnitude curve for each fault type, wherein the height of the curve represents the magnitude of the probability of diagnosing the fault type, and a higher curve indicates a greater probability that the unknown signal is a fault of the type. It can be seen from fig. 14 to 21 that the AVMD-correlation analysis method has no cross over on the diagnosis result curves of various fault types, and the diagnosis effect is good. As shown in table 2, the VMD-correlation analysis algorithm, EMD-correlation algorithm, EEMD-correlation algorithm, wavelet decomposition-correlation analysis algorithm were used for the fault diagnosis, respectively. Wherein the VMD-correlation analysis algorithm is very fast and efficient. Although the wavelet decomposition-correlation analysis algorithm is fast in diagnosis speed, the wavelet decomposition has obvious defects on the frequency domain decomposition effect of the signal as can be seen from fig. 1 and 4. Therefore, the method has strong advantages in decomposition effect and calculation efficiency.
TABLE 2 comparison of operating times of four algorithms
Claims (1)
1. A wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis is characterized in that: the method comprises two stages of 'fault library modeling' and 'implementation diagnosis', and comprises the following specific steps:
A. establishing a fault library stage:
1) a fault signal decomposition stage;
decomposing the fault signal into N modal components by using an adaptive variational modal decomposition method AVMD, and repeatedly decomposing each fault signal by using the AVMD for M types of fault signals for M times; obtaining M groups of signal component sets with the number of N;
2) a characteristic spectrum acquisition stage;
firstly, carrying out Principal Component Analysis (PCA) denoising and dimensionality reduction on modal components of M fault signals, selecting the first n principal component components with the highest contribution rate, and converting the principal component components into frequency spectrums by using Fast Fourier Transform (FFT); connecting the frequency spectrums of the n principal component components end to form a fault characteristic pseudo frequency spectrum; for M types of fault signals, a fault characteristic pseudo-spectrum vector needs to be constructed for each type of fault signal; a set A containing the M characteristic pseudo-spectrum vectors is called a fault characteristic set;
B. the diagnosis phase is carried out:
decomposing unknown fault signals into N intrinsic mode components by using an AVMD method, reducing and denoising the N intrinsic mode components by using PCA, converting the N intrinsic mode components into N principal component components, converting the first N principal component components with the highest contribution rate into frequency spectrums by using fast Fourier transform, and forming pseudo-spectrum characteristic vectors by connecting the N frequency spectrums end to end; respectively solving cross correlation coefficients of the pseudo-spectrum feature vectors and all fault pseudo-spectrum feature vectors in the fault feature set A, wherein the fault type corresponding to the maximum cross correlation coefficient is a diagnosis result; the cross-correlation coefficient r is calculated as follows:
wherein X represents a pseudo-spectrum feature vector of an unknown fault signal, and Y represents a pseudo-spectrum feature vector of one fault type in a fault feature set; cov represents the covariance of two pseudo-spectrum feature vectors; σ represents the standard deviation;
determining the number of the modes; before decomposing the signal variation mode, drawing a short-time Fourier transform spectrogram by using short-time Fourier transform, and acquiring the mode number N according to an image;
determining a penalty parameter epsilon; changing penalty parameters in variation modal decomposition by adopting a correction step length tau, merging modal results after each decomposition, and solving a cross-correlation coefficient with an original signal; obtaining an optimal penalty parameter epsilon according to an inflection point of the cross-correlation coefficient curve;
decomposing the fault signal using the AVMD; and setting AVMD parameters according to the determined number N of the eigenmodes and the optimal penalty parameter epsilon, and then decomposing the fault signal by using the AVMD to obtain component signals of different modes.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811507395.9A CN109655266B (en) | 2018-12-11 | 2018-12-11 | Wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811507395.9A CN109655266B (en) | 2018-12-11 | 2018-12-11 | Wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109655266A CN109655266A (en) | 2019-04-19 |
CN109655266B true CN109655266B (en) | 2021-12-10 |
Family
ID=66113254
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811507395.9A Active CN109655266B (en) | 2018-12-11 | 2018-12-11 | Wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109655266B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111060301B (en) * | 2019-12-27 | 2021-11-19 | 中国联合网络通信集团有限公司 | Fault diagnosis method and device |
CN112113766B (en) * | 2020-09-01 | 2021-11-09 | 兰州理工大学 | Characteristic extraction method for early damage state of rolling bearing |
CN112798280B (en) * | 2021-02-05 | 2022-01-04 | 山东大学 | Rolling bearing fault diagnosis method and system |
CN112835104A (en) * | 2021-03-26 | 2021-05-25 | 中国石油大学(华东) | Unconventional reservoir natural frequency in-situ measurement system |
US11539317B2 (en) | 2021-04-05 | 2022-12-27 | General Electric Renovables Espana, S.L. | System and method for detecting degradation in wind turbine generator bearings |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107192554A (en) * | 2017-05-19 | 2017-09-22 | 西安理工大学 | A kind of vibrating failure diagnosis method of Wind turbines rolling bearing |
CN107832525A (en) * | 2017-11-07 | 2018-03-23 | 昆明理工大学 | A kind of method and its application of information entropy optimization VMD extractions bearing fault characteristics frequency |
CN108387373A (en) * | 2017-12-06 | 2018-08-10 | 上海电力学院 | The Fault Diagnosis of Roller Bearings of variation mode decomposition is improved based on related coefficient |
-
2018
- 2018-12-11 CN CN201811507395.9A patent/CN109655266B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107192554A (en) * | 2017-05-19 | 2017-09-22 | 西安理工大学 | A kind of vibrating failure diagnosis method of Wind turbines rolling bearing |
CN107832525A (en) * | 2017-11-07 | 2018-03-23 | 昆明理工大学 | A kind of method and its application of information entropy optimization VMD extractions bearing fault characteristics frequency |
CN108387373A (en) * | 2017-12-06 | 2018-08-10 | 上海电力学院 | The Fault Diagnosis of Roller Bearings of variation mode decomposition is improved based on related coefficient |
Non-Patent Citations (4)
Title |
---|
基于提高变分模态分解的齿轮箱复合故障特征提取;柴慧理 等;《机械传动》;20180731;第42卷(第7期);全文 * |
基于自适应数学形态学的滚动轴承故障诊断方法;齐咏生 等;《大连理工大学学报》;20180531;第58卷(第3期);第238-245页 * |
广义变分模态分解方法及其在变工况齿轮故障诊断中的应用;郑近德 等;《振动工程学报》;20170630;第30卷(第3期);第502-509页 * |
盲VMD-Cepstral在轴承故障诊断中的应用;柏林 等;《振动、测试与诊断》;20180630;第38卷(第3期);第597-602页 * |
Also Published As
Publication number | Publication date |
---|---|
CN109655266A (en) | 2019-04-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109655266B (en) | Wind turbine generator bearing fault diagnosis method based on AVMD and spectrum correlation analysis | |
Wang et al. | Fault diagnosis of a rolling bearing using wavelet packet denoising and random forests | |
CN109827777B (en) | Rolling bearing fault prediction method based on partial least square method extreme learning machine | |
Yang et al. | Vibration feature extraction techniques for fault diagnosis of rotating machinery: a literature survey | |
CN109000921B (en) | Method for diagnosing main shaft fault of wind turbine generator | |
CN109255333B (en) | Fault hybrid diagnosis method for rolling bearing of large wind turbine generator | |
CN108388908B (en) | Rolling bearing impact fault diagnosis method based on K-SVD and sliding window noise reduction | |
CN105784353A (en) | Fault diagnosis method for gear case of aerogenerator | |
Sawalhi et al. | Spectral kurtosis optimization for rolling element bearings. | |
CN110595778B (en) | Wind turbine generator bearing fault diagnosis method based on MMF and IGRA | |
Ding et al. | Sparsity-based algorithm for condition assessment of rotating machinery using internal encoder data | |
CN109883706A (en) | A kind of rolling bearing local damage Weak fault feature extracting method | |
CN111120388B (en) | Fan state combined monitoring method and system | |
CN113569990B (en) | Strong noise interference environment-oriented performance equipment fault diagnosis model construction method | |
Huang et al. | An improved empirical wavelet transform method for rolling bearing fault diagnosis | |
CN114112400A (en) | Mechanical bearing fault diagnosis method based on multi-angle information fusion | |
CN115434872A (en) | Wind turbine generator gearbox composite fault diagnosis method based on AVMD and improved RSSD | |
CN113250911B (en) | Fan blade fault diagnosis method based on VMD decomposition algorithm | |
CN117109923A (en) | Rolling bearing fault diagnosis method and system | |
CN110646199B (en) | Gear weak fault diagnosis method based on weighted derivative dynamic time warping | |
CN112177865A (en) | Method for solving marking noise and insufficient marks in fan fault detection | |
CN116378912A (en) | Early compound fault diagnosis method and device for wind driven generator and readable storage medium | |
Zhang et al. | Weighted sparsity-based denoising for extracting incipient fault in rolling bearing | |
CN113686577B (en) | Bearing fault diagnosis method based on rapid nonlinear sparse spectrum | |
CN113758708B (en) | Rolling bearing signal frequency domain fault diagnosis method based on L1 norm and group norm constraint |
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 |