CN110987433A - Bearing fault early warning method based on high-frequency signal characteristic amplitude - Google Patents
Bearing fault early warning method based on high-frequency signal characteristic amplitude Download PDFInfo
- Publication number
- CN110987433A CN110987433A CN201911280815.9A CN201911280815A CN110987433A CN 110987433 A CN110987433 A CN 110987433A CN 201911280815 A CN201911280815 A CN 201911280815A CN 110987433 A CN110987433 A CN 110987433A
- Authority
- CN
- China
- Prior art keywords
- fault
- bearing
- early warning
- stage
- amplitude
- 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
Landscapes
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a bearing fault early warning method based on high-frequency signal characteristic amplitude, which belongs to the field of early warning of faults of rolling bearings and comprises the following steps: collecting vibration signals of a training bearing at equal intervals; performing discrete wavelet transform on the vibration signal and extracting high-frequency components; sequencing the absolute values of the high-frequency components to obtain an enhanced impact amplitude LR and a carpet impact amplitude HR; taking an LR-HR value corresponding to the same moment as the early warning characteristic of the moment; dividing the normal stage and the fault stage of the training bearing according to the LR-HR value of the whole life cycle of the training bearing; respectively carrying out Weibull distribution fitting on the early warning characteristics of each stage to obtain probability distribution; collecting an LR-HR value of a test bearing in operation; and calculating the probability of the test bearing in a normal stage and a fault stage based on the probability distribution. The invention utilizes the high-frequency noise signal to carry out fault early warning, avoids complex denoising and signal enhancement processes, has high early warning speed and is suitable for practical application, and the fault early warning characteristic can be increased along with the degradation severity.
Description
Technical Field
The invention belongs to the field of rolling bearing running state monitoring and fault early warning, and particularly relates to a bearing fault early warning method based on high-frequency signal characteristic amplitude.
Background
Rolling bearings are widely used in the industrial field and are known as industrial joints. Whether the rolling bearing can normally operate relates to the production quality and the personnel and property safety of the whole industrial production process. In the operation process, the rolling bearing is degraded along with the operation time due to environmental factors, wear factors, human factors and the like, and faults of the inner ring, the outer ring and the rotor part can be generated after the rolling bearing is accumulated for a certain time. In order to avoid economic and safety problems caused by major faults of the bearing, the online monitoring and early warning of the rolling bearing are strongly necessary, and the online monitoring and early warning are the basis for ensuring the safe operation of the whole industrial production process.
The most common rolling bearing fault detection techniques currently used include: a method based on signal processing and a method of staging. The method based on signal processing mainly aims at searching a characteristic frequency band of fault occurrence and harmonic waves of fault characteristic frequency, and the processing steps comprise: denoising by adopting methods such as DRS (Discrete random separation), TSA (time synchronous averaging) and the like, then performing signal enhancement by using methods such as a peak maximum value, minimum entropy deconvolution and the like, and extracting the characteristic frequency of the fault; when the bearing is seriously degraded, strong jitter is often accompanied, strong background noise exists, the fault characteristic frequency of the bearing is easily swallowed, and the method is difficult to extract the fault characteristic frequency. The stage division method divides normal and fault stages of the bearing through off-line modeling, comprises the steps of extracting degradation characteristics RMS (Root mean square), Kurtosis (Kurtosis) and the like of the bearing, then artificially defining a normal stage, defining a threshold value for the degradation characteristics of the health stage by using a method such as a 3 sigma principle or a Chebyshev inequality, and starting early warning when the characteristic value exceeds the threshold value. The characteristic fluctuation extracted by the method is large, the false alarm rate of early warning is high, and the early warning is unreliable.
Generally, the existing rolling bearing fault detection method has the problems that the fault characteristic frequency is difficult to detect when the bearing is seriously degraded, the fluctuation of the extracted degradation characteristic is high, and the false alarm rate is high.
Disclosure of Invention
Aiming at the defects or improvement requirements of the prior art, the invention provides a bearing fault early warning method based on a high-frequency signal characteristic amplitude, and aims to solve the technical problems that the existing rolling bearing fault detection method cannot detect fault characteristic frequency when a bearing is seriously degraded, the fluctuation of extracted degradation characteristics is high, and the false alarm rate is high.
In order to achieve the above object, according to an aspect of the present invention, there is provided a bearing fault early warning method based on a high-frequency signal characteristic amplitude, including:
s1: sampling the vibration signals of the whole life cycle of the training bearing at equal intervals;
s2: performing discrete wavelet transform on the sampled vibration signals, and performing high-frequency component extraction on the transformed signals;
s3: sorting the absolute values of the high-frequency components corresponding to the same sampling moment from large to small, taking the absolute value arranged at a first set position as an enhanced impact amplitude LR, taking the absolute value arranged at a second set position as a carpet impact amplitude HR, and taking the difference value between the enhanced impact amplitude LR and the carpet impact amplitude HR as the fault early warning characteristic at the moment;
s4: dividing the whole life cycle of the training bearing into a normal stage and a fault stage according to the fault early warning characteristics corresponding to all sampling moments;
s5: respectively carrying out Weibull distribution fitting on the fault early warning characteristics of the normal stage and the fault stage to obtain probability density distribution of the normal stage and the fault stage;
s6: acquiring a vibration signal of the bearing to be pre-warned at the current moment, and obtaining the fault pre-warning characteristics of the bearing to be pre-warned at the current moment according to the steps S2-S3;
s7: and respectively calculating probability values of the fault early warning characteristics at the current moment belonging to a normal stage and a fault stage according to the probability density distribution obtained in the step S5 to obtain a fault early warning result of the bearing to be early warned at the current moment.
Further, in step S2, the discrete wavelet transform uses a Harr wavelet basis, and the high-pass filter of the Harr wavelet basis is:
wherein, H (x)kThe kth value, a, representing the high frequency output2kRepresenting the 2 k-th value of the sampled vibration signal.
Further, the intensified shock amplitude LR in step S3 is an absolute value sorted in the range of twenty to thirty percent; the carpet impact amplitude HR is an absolute value ordered at seventy to eighty percent.
Further, step S4 divides the training bearing into a normal stage and a fault stage according to the fault early warning features corresponding to all sampling moments, specifically: and finding out the positions of the catastrophe points in the early warning features corresponding to all sampling moments of the training bearing, wherein the catastrophe points are in a normal stage before and in a fault stage after.
Further, in step S5, weibull distribution fitting is performed on the fault early warning features in the normal stage and the fault stage, specifically, weibull distribution fitting is performed on the fault early warning features in the normal stage and the fault stage by using maximum likelihood estimation.
Further, step S7 specifically includes:
s7.1: respectively substituting the fault early-warning characteristics of the bearing to be early-warned at the current moment into the probability density distribution of the normal stage and the fault stage and integrating to obtain the normal probability P (N) and the fault probability P (F) of the bearing to be early-warned;
s7.2: comparing the normal probability P (N) and the fault probability P (F) of the bearing to be early-warned, and performing fault early warning when P (F) is greater than P (N); otherwise, the bearing to be pre-warned is in a normal stage.
According to another aspect of the present invention, a computer-readable storage medium is provided, in which a computer program is stored, and the computer program is executed by a processor to implement the above-mentioned bearing fault early warning method based on high-frequency signal characteristic amplitude.
In general, compared with the prior art, the above technical solution contemplated by the present invention can achieve the following beneficial effects:
(1) according to the bearing fault early warning method provided by the invention, the characteristic that the high-frequency noise signal becomes larger along with the larger fault when the fault occurs in the bearing is utilized, the high-frequency signal extraction and the amplitude analysis are carried out on the vibration signal in the bearing operation process, so that the fault early warning can be effectively carried out, the complicated steps of denoising and signal enhancement are avoided, and the real-time analysis decision of the industrial process is facilitated.
(2) According to the bearing early warning method provided by the invention, Weibull distribution fitting is respectively carried out on the early warning characteristics of the bearing in the normal stage and the fault stage to obtain the probability density distribution of the normal stage and the fault stage, the bearing is monitored by utilizing the probability density distribution, and a more accurate early warning result is obtained.
(3) Because the method of the invention utilizes the high-frequency noise signal to carry out the fault early warning, the fault early warning characteristic can not be interfered by enhanced background noise even when the bearing is seriously degraded, and the fault early warning characteristic can be increased along with the degradation severity degree, thereby being more suitable for practical application.
Drawings
FIG. 1 is a flow chart of a bearing fault early warning method based on high-frequency signal characteristic amplitude according to an embodiment of the present invention;
FIG. 2 is a wave diagram of a vibration signal acquired by a training bearing at a certain sampling time point;
FIG. 3(a) is a diagram illustrating the result of extracting high frequency components after performing discrete wavelet transform on the acquired vibration signals;
FIG. 3(b) is a diagram showing the absolute value of the high frequency component;
FIG. 4 is a graphical representation of carpet impact (HR) amplitude and reinforcement impact (LR) amplitude taken for the absolute value of the high frequency component;
FIG. 5 is a schematic of a training bearing full life cycle LR-HR index;
FIG. 6(a) is a schematic of a Weibull distribution fit to LR-HR for a normal phase of training a bearing;
FIG. 6(b) is a schematic of a Weibull distribution fit to LR-HR during a training bearing fault phase;
FIG. 7(a) is a schematic of a test bearing full life cycle LR-HR;
FIG. 7(b) is a schematic partially developed view of a test bearing at the point of failure occurrence;
FIG. 8 is a graphical representation of the normal Probability (PN) and failure Probability (PF) of a test bearing as a function of time.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
In order to verify the effectiveness of the bearing fault early warning method based on the high-frequency signal characteristic amplitude and facilitate understanding of the technical scheme, the embodiment of the invention adopts the method to carry out fault early warning test on the bearing data of the FEMTO test bed. The FEMTO test bed comprises a rotating motor, a transmission mechanism, a test bearing and a load device, and the test bearing is accelerated to age in a load overload mode so as to break down and damage. Through setting up rotational speed and load, divide into three operating mode with the simulation operational environment of difference with the experiment: wherein, the working condition one has two training bearings 1-1 and 1-2 and five testing bearings 1-3 to 1-7. The present embodiment will explain the method of the present invention in detail by taking the first operating condition as an example. As shown in fig. 1, the bearing fault early warning method based on the characteristic amplitude of the high-frequency signal provided by the invention comprises the following steps:
s1: sampling the vibration signals of the whole life cycle of the training bearing at equal intervals;
specifically, a section of vibration data collected for the training bearing 1-1 by the embodiment of the present invention is shown in fig. 2.
S2: performing discrete wavelet transform on the sampled vibration signals, and performing high-frequency component extraction on the transformed signals;
specifically, when a fault occurs in the bearing, the characteristic frequency and the harmonic wave of the fault generally occur in a high-frequency band, a high-frequency band signal contains effective fault information, and a high-frequency noise signal becomes larger as the fault becomes larger, so that the high-frequency signal extraction and amplitude analysis are performed on a vibration signal in the running process of the bearing, so that the fault can be effectively detected in the following process. Therefore, the invention adopts discrete wavelet transform to extract high-frequency signals. The wavelet base of the discrete wavelet transform comprises a Harr wavelet, a db wavelet, a sym wavelet and the like, wherein the Harr wavelet is relatively simple and easy to realize, so that the Harr wavelet base is selected to construct a high-pass filter in the embodiment of the invention:
wherein, H (x)kThe kth value, a, representing the high frequency output2kRepresenting the 2 k-th value of the sampled vibration signal.
The high frequency component obtained by performing discrete wavelet transform on the training set bearing is shown in fig. 3(a), and it can be seen that the length of the high frequency component obtained after performing discrete wavelet transform is reduced by half, and the high frequency includes noise components and some higher harmonic components of vibration signals. The result of taking the absolute value of the high frequency component is shown in fig. 3(b), and the absolute value is taken to better analyze the amplitude and remove the influence of the sign of the signal.
S3: sorting the absolute values of the high-frequency components corresponding to the same sampling moment from large to small, taking the absolute value arranged at a first set position as an enhanced impact amplitude LR, taking the absolute value arranged at a second set position as a carpet impact amplitude HR, and taking the difference value between the enhanced impact amplitude LR and the carpet impact amplitude HR as the fault early warning characteristic at the moment;
specifically, HR represents the carpet impact amplitude, characterizing the intensity achieved by most amplitudes; LR indicates the intensified shock amplitude, and is a relatively large shock amplitude intensity achieved by a small number of signals. When a fault occurs, fault pulses are generated in the high-frequency signal, LR as the extracted impact amplitude can be increased along with the increase of the fault pulses, HR is approximately constant, and therefore LR-HR can represent the size of the fault. LR and HR are taken as the amplitude representatives of the sampling signals at the moment, and the effect of reducing the dimension is achieved. The change of LR and HR with time can represent the information of the change of the intensity of the fault pulse with time. According to the embodiment of the invention, the pulse generation frequency and proportion are strengthened according to experimental data, and the absolute value sequenced at the first twenty-five percent is used as the strengthened impact amplitude LR; the absolute value ordered at the top seventy-five percent is taken as the carpet impact amplitude HR, and in practical application, LR may be selected to be twenty to thirty percent and HR may be selected to be eighty to seventy percent, depending on the frequency and proportion of strengthening pulses generated when a failure occurs in an actual running bearing. The results of the carpet impact amplitude HR and the reinforcement impact amplitude LR of the high-frequency component absolute value corresponding to the same sampling time in the embodiment of the present invention are shown in fig. 4, and the corresponding early warning feature at this time is (LR-HR).
S4: dividing the training bearing into a normal stage and a fault stage according to the fault early warning characteristics corresponding to all sampling moments;
specifically, the (LR-HR) values at all sampling times are plotted against time as shown in fig. 5, and the (LR-HR) value is relatively small in the early stage, which is the normal stage of healthy operation; and the later period (LR-HR) is obviously increased, which indicates that the bearing has a fault, the running state has a problem and needs early warning in time.
The fault early warning characteristics of the training bearing can generate mutation when a fault occurs, and the position of a mutation point is found out from the early warning characteristics of the training bearing at all sampling moments. The catastrophe point is preceded by a normal phase and is followed by a fault phase. As shown in fig. 5, the discontinuity of the bearing 1-1 occurs at a 2460 time point.
S5: respectively carrying out Weibull distribution fitting on the fault early warning characteristics of the normal stage and the fault stage to obtain probability density distribution of the normal stage and the fault stage;
particularly, the Weibull distribution is widely applied to the reliability field, has stronger fitting capacity on bearing degradation and fault data, and obtains more accurate distribution. The maximum likelihood estimation is used as a common parameter estimation method, and can accurately estimate parameters of Weibull distribution, so that the embodiment of the invention adopts the maximum likelihood estimation to respectively carry out Weibull distribution fitting on fault early warning characteristics of a normal stage and a fault stage. The result of weibull distribution fitting to the LR-HR at the normal stage of the training bearing is shown in fig. 6(a), and the result of weibull distribution fitting to the LR-HR at the fault stage of the training bearing is shown in fig. 6 (b); the distribution of the normal samples and the distribution of the fault samples can be seen from the distribution of the two stages, so that the distribution of the normal samples and the distribution of the fault samples have good discrimination, and the bearing of the test set can be subjected to fault detection and early warning by using two probability densities.
S6: acquiring a vibration signal of the bearing to be pre-warned at the current moment, and obtaining the fault pre-warning characteristics of the bearing to be pre-warned at the current moment according to the steps S2-S3;
specifically, fig. 7(a) and 7(b) are partial enlarged views of the variation curve of the LR-HR full life cycle of the test bearing and the occurrence time of the fault, respectively, and it can be seen from fig. 7(b) that the fault occurs, that is, the time point at which the LR-HR undergoes sudden change is about at 2250 time.
S7: and respectively calculating probability values of the fault early warning characteristics at the current moment belonging to a normal stage and a fault stage according to the probability density distribution obtained in the step S5 to obtain a fault early warning result of the bearing to be early warned at the current moment.
Specifically, the change curves of the normal probability p (n) and the failure probability p (f) of the test bearing along with time are shown in fig. 8, the normal probability p (n) of the initial healthy operation is much greater than the failure probability p (f), and the failure probability is substantially 0; when the fault occurs, the normal probability P (N) is reduced, and the fault probability P (F) is rapidly increased; the moment when the failure probability p (f) is greater than the normal probability for the first time is 2251, it can be seen that the result is consistent with the result observed in fig. 7(b), which proves that the method of the present invention can rapidly give an early warning when a failure occurs.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (7)
1. A bearing fault early warning method based on high-frequency signal characteristic amplitude is characterized by comprising the following steps:
s1: sampling the vibration signals of the whole life cycle of the training bearing at equal intervals;
s2: performing discrete wavelet transform on the sampled vibration signals, and performing high-frequency component extraction on the transformed signals;
s3: sorting the absolute values of the high-frequency components corresponding to the same sampling moment from large to small, taking the absolute value arranged at a first set position as an enhanced impact amplitude LR, taking the absolute value arranged at a second set position as a carpet impact amplitude HR, and taking the difference value between the enhanced impact amplitude LR and the carpet impact amplitude HR as the fault early warning characteristic at the moment;
s4: dividing the whole life cycle of the training bearing into a normal stage and a fault stage according to the fault early warning characteristics corresponding to all sampling moments;
s5: respectively carrying out Weibull distribution fitting on the fault early warning characteristics of the normal stage and the fault stage to obtain probability density distribution of the normal stage and the fault stage;
s6: acquiring a vibration signal of the bearing to be pre-warned at the current moment, and obtaining the fault pre-warning characteristics of the bearing to be pre-warned at the current moment according to the steps S2-S3;
s7: and respectively calculating probability values of the fault early warning characteristics at the current moment belonging to a normal stage and a fault stage according to the probability density distribution obtained in the step S5 to obtain a fault early warning result of the bearing to be early warned at the current moment.
2. The bearing fault early warning method based on the high-frequency signal characteristic amplitude as claimed in claim 1, wherein the discrete wavelet transform in step S2 adopts Harr wavelet basis, and the high-pass filter of the Harr wavelet basis is:
wherein, H (x)kThe kth value, a, representing the high frequency output2kRepresenting the 2 k-th value of the sampled vibration signal.
3. The bearing fault early warning method based on the high-frequency signal characteristic amplitude as claimed in claim 1 or 2, wherein the intensified shock amplitude LR in step S3 is an absolute value sorted in a range of twenty to thirty percent; the carpet impact amplitude HR is an absolute value ordered at seventy to eighty percent.
4. The bearing fault early warning method based on the high-frequency signal characteristic amplitude as claimed in any one of claims 1 to 3, wherein in step S4, according to the fault early warning characteristics corresponding to all sampling moments, the training bearing is divided into a normal stage and a fault stage, specifically:
and finding out the positions of the catastrophe points in the early warning features corresponding to all sampling moments of the training bearing, wherein the catastrophe points are in a normal stage before and in a fault stage after.
5. The bearing fault early warning method based on the high-frequency signal characteristic amplitude as claimed in any one of claims 1 to 4, wherein step S5 is performed with Weibull distribution fitting on the fault early warning characteristics in the normal stage and the fault stage respectively, specifically, maximum likelihood estimation is performed with Weibull distribution fitting on the fault early warning characteristics in the normal stage and the fault stage respectively.
6. The bearing fault early warning method based on the high-frequency signal characteristic amplitude as claimed in any one of claims 1 to 5, wherein the step S7 specifically comprises:
s7.1: respectively substituting the fault early-warning characteristics of the bearing to be early-warned at the current moment into the probability density distribution of the normal stage and the fault stage and integrating to obtain the normal probability P (N) and the fault probability P (F) of the bearing to be early-warned;
s7.2: comparing the normal probability P (N) and the fault probability P (F) of the bearing to be early-warned, and performing fault early warning when P (F) is greater than P (N); otherwise, the bearing to be pre-warned is in a normal stage.
7. A computer-readable storage medium, wherein the computer-readable storage medium stores thereon a computer program, and when the computer program is executed by a processor, the method for bearing fault pre-warning based on high-frequency signal characteristic amplitude as claimed in any one of claims 1 to 6 is implemented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911280815.9A CN110987433B (en) | 2019-12-13 | 2019-12-13 | Bearing fault early warning method based on high-frequency signal characteristic amplitude |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911280815.9A CN110987433B (en) | 2019-12-13 | 2019-12-13 | Bearing fault early warning method based on high-frequency signal characteristic amplitude |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110987433A true CN110987433A (en) | 2020-04-10 |
| CN110987433B CN110987433B (en) | 2021-05-18 |
Family
ID=70093278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911280815.9A Expired - Fee Related CN110987433B (en) | 2019-12-13 | 2019-12-13 | Bearing fault early warning method based on high-frequency signal characteristic amplitude |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110987433B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111811819A (en) * | 2020-06-30 | 2020-10-23 | 佛山科学技术学院 | Bearing fault diagnosis method and device based on machine learning |
| CN112098094A (en) * | 2020-09-27 | 2020-12-18 | 上海数深智能科技有限公司 | Method for diagnosing fault vibration of low-speed heavy-load bearing |
| CN115688612A (en) * | 2023-01-03 | 2023-02-03 | 北京博华信智科技股份有限公司 | Method, device, equipment and medium for predicting life of rolling bearing |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103837345A (en) * | 2014-03-25 | 2014-06-04 | 上海电机学院 | Gearbox fault diagnosis method and device |
| CN103954450A (en) * | 2014-05-19 | 2014-07-30 | 重庆交通大学 | Bearing life degradation performance evaluation index construction method based on main component analysis |
| JP2014206403A (en) * | 2013-04-11 | 2014-10-30 | Ntn株式会社 | Rolling bearing diagnostic system |
| CN104849590A (en) * | 2015-05-06 | 2015-08-19 | 中国人民解放军军械工程学院 | Method for detecting weak pulse signals under mixed noise interference |
| CN108181105A (en) * | 2017-11-28 | 2018-06-19 | 杭州安脉盛智能技术有限公司 | Logic-based returns and the rolling bearing fault of J divergences examines method and system in advance |
| CN109946075A (en) * | 2018-12-25 | 2019-06-28 | 东北大学 | A kind of bearing condition monitoring and method for diagnosing faults |
| CN110231160A (en) * | 2019-07-11 | 2019-09-13 | 电子科技大学中山学院 | Method and device for predicting service life of mechanical part of electric drive assembly |
-
2019
- 2019-12-13 CN CN201911280815.9A patent/CN110987433B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014206403A (en) * | 2013-04-11 | 2014-10-30 | Ntn株式会社 | Rolling bearing diagnostic system |
| CN103837345A (en) * | 2014-03-25 | 2014-06-04 | 上海电机学院 | Gearbox fault diagnosis method and device |
| CN103954450A (en) * | 2014-05-19 | 2014-07-30 | 重庆交通大学 | Bearing life degradation performance evaluation index construction method based on main component analysis |
| CN104849590A (en) * | 2015-05-06 | 2015-08-19 | 中国人民解放军军械工程学院 | Method for detecting weak pulse signals under mixed noise interference |
| CN108181105A (en) * | 2017-11-28 | 2018-06-19 | 杭州安脉盛智能技术有限公司 | Logic-based returns and the rolling bearing fault of J divergences examines method and system in advance |
| CN109946075A (en) * | 2018-12-25 | 2019-06-28 | 东北大学 | A kind of bearing condition monitoring and method for diagnosing faults |
| CN110231160A (en) * | 2019-07-11 | 2019-09-13 | 电子科技大学中山学院 | Method and device for predicting service life of mechanical part of electric drive assembly |
Non-Patent Citations (1)
| Title |
|---|
| 姜小荧: "基于小波分析的滚动轴承故障诊断方法的研究及应用", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111811819A (en) * | 2020-06-30 | 2020-10-23 | 佛山科学技术学院 | Bearing fault diagnosis method and device based on machine learning |
| CN112098094A (en) * | 2020-09-27 | 2020-12-18 | 上海数深智能科技有限公司 | Method for diagnosing fault vibration of low-speed heavy-load bearing |
| CN115688612A (en) * | 2023-01-03 | 2023-02-03 | 北京博华信智科技股份有限公司 | Method, device, equipment and medium for predicting life of rolling bearing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110987433B (en) | 2021-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110987433B (en) | Bearing fault early warning method based on high-frequency signal characteristic amplitude | |
| CN108760327B (en) | Diagnosis method for rotor fault of aircraft engine | |
| CN108804832B (en) | VMD-based interval threshold partial discharge denoising method | |
| Li et al. | Early fault diagnosis of rotating machinery by combining differential rational spline-based LMD and K–L divergence | |
| CN112101174A (en) | LOF-Kurtogram-based mechanical fault diagnosis method | |
| CN111238843B (en) | Fan health evaluation method based on rapid spectrum kurtosis analysis | |
| CN109190598B (en) | Rotating machinery monitoring data noise point detection method based on SES-LOF | |
| Meng et al. | Health indicator of bearing constructed by rms-CUMSUM and GRRMD-CUMSUM with multifeatures of envelope spectrum | |
| Wang et al. | Fault diagnosis of wind turbine bearing based on optimized adaptive chirp mode decomposition | |
| Chen et al. | Integrated early fault diagnosis method based on direct fast iterative filtering decomposition and effective weighted sparseness kurtosis to rolling bearings | |
| CN112487882B (en) | Method for generating non-sparse index-guided enhanced envelope spectrum based on spectrum coherence | |
| CN110717472A (en) | Fault diagnosis method and system based on improved wavelet threshold denoising | |
| Zhang et al. | Improved local cepstrum and its applications for gearbox and rolling bearing fault detection | |
| CN109596354B (en) | Band-pass filtering method based on self-adaptive resonance frequency band identification | |
| CN110956112B (en) | Novel high-reliability slewing bearing service life assessment method | |
| CN108507782A (en) | The detection method of periodic signal implicit cycle under a kind of strong background noise | |
| CN109872511B (en) | Self-adaptive two-stage alarm method for monitoring axial displacement sudden change | |
| CN111380680A (en) | Check valve fault feature extraction method based on improved permutation entropy | |
| CN109828548A (en) | Performance degradation feature evaluation method based on time series variation Singularity detection | |
| CN112903296B (en) | Rolling bearing fault detection method and system | |
| CN112036270B (en) | Vibration monitoring method for early damage state of rolling bearing | |
| CN109446625B (en) | Bayesian inference-based helicopter maneuvering component dynamic threshold calculation method | |
| Wang et al. | A fault diagnosis method for automaton based on morphological component analysis and ensemble empirical mode decomposition | |
| Klein et al. | Methods for bearing feature extraction | |
| CN107884188B (en) | A kind of variable speed/varying load Fault Diagnosis of Roller Bearings based on mode conversion |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210518 Termination date: 20211213 |