CN112945557A - Slewing bearing fault diagnosis method and device and storage medium - Google Patents
Slewing bearing fault diagnosis method and device and storage medium Download PDFInfo
- Publication number
- CN112945557A CN112945557A CN202110130106.3A CN202110130106A CN112945557A CN 112945557 A CN112945557 A CN 112945557A CN 202110130106 A CN202110130106 A CN 202110130106A CN 112945557 A CN112945557 A CN 112945557A
- Authority
- CN
- China
- Prior art keywords
- slewing bearing
- signal
- optimal
- fault
- imf
- 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000003745 diagnosis Methods 0.000 title claims abstract description 38
- 238000003860 storage Methods 0.000 title claims abstract description 12
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 35
- 238000012545 processing Methods 0.000 claims abstract description 33
- 208000036075 Autosomal dominant tubulointerstitial kidney disease Diseases 0.000 claims abstract description 30
- 101001133056 Homo sapiens Mucin-1 Proteins 0.000 claims abstract description 29
- 102100034256 Mucin-1 Human genes 0.000 claims abstract description 29
- 238000005457 optimization Methods 0.000 claims abstract description 24
- 238000001228 spectrum Methods 0.000 claims abstract description 9
- 238000010183 spectrum analysis Methods 0.000 claims abstract description 9
- 241000282461 Canis lupus Species 0.000 claims description 37
- 238000000354 decomposition reaction Methods 0.000 claims description 24
- 230000001133 acceleration Effects 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 9
- 238000012935 Averaging Methods 0.000 claims description 5
- 238000002474 experimental method Methods 0.000 claims description 5
- 238000013178 mathematical model Methods 0.000 claims description 3
- 238000010835 comparative analysis Methods 0.000 claims description 2
- 238000011156 evaluation Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000005096 rolling process Methods 0.000 description 6
- 238000007792 addition Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000002405 diagnostic procedure Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 208000028804 PERCHING syndrome Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000036528 appetite Effects 0.000 description 1
- 235000019789 appetite Nutrition 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/15—Correlation function computation including computation of convolution operations
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Acoustics & Sound (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a fault diagnosis method, a fault diagnosis device and a storage medium for a slewing bearing, wherein the method comprises the following steps: s1, acquiring a vibration signal of the slewing bearing, and setting the vibration signal as an original signal; s2, processing the original signal by an EEMD method, decomposing to obtain an intrinsic mode component IMF of the vibration signal, and then performing preferential processing on the IMF to obtain an optimal component; s3, carrying out MCKD parameter optimization processing on the optimal component by using GWO algorithm and taking the correlation kurtosis as a fitness function to obtain an optimal parameter combination; s4, substituting the optimal parameter combination obtained by the optimization processing of the GWO algorithm into the MCKD to analyze the optimal component to obtain a noise reduction signal; s5, carrying out envelope spectrum analysis on the noise reduction signal, and then comparing and analyzing the fault characteristic frequency found in the envelope spectrum with the theoretical fault characteristic frequency to obtain a diagnosis result.
Description
Technical Field
The invention relates to the technical field of bearing detection, in particular to a method and a device for diagnosing faults of a slewing bearing and a storage medium.
Background
The slewing bearing is a special large-scale rolling bearing and is widely applied to various large-scale machines, such as a portal crane of a port. The portal crane slewing bearing has uneven stress and complex motion state of each part when in working state, and due to the running characteristics of low speed and heavy load and a large amount of background noise, the original impact generated by the fault is submerged in the background noise, so that the extraction of fault characteristics is more difficult than that of a common medium-high speed bearing. However, in the long-term use of the portal crane, the occurrence of a failure of the slewing bearing is inevitable. If the faults can not be found and countermeasures can not be taken in time, the normal operation of the gantry crane can be influenced, and great economic loss is caused.
At present, a small number of researchers develop research on fault diagnosis of the slewing bearing and obtain preliminary results. Lv's Showa et al uses a composite method of wavelet decomposition and extreme mean modal decomposition to diagnose faults of a portal crane slewing bearing. However, the wavelet base and the decomposition layer number in the wavelet decomposition cannot be selected adaptively. The Nanjing university of industry makes more researches on the fault diagnosis of the slewing bearing, applies methods such as circular domain analysis, wavelet energy mode, bispectrum analysis, kurtosis probability density analysis, weighted fusion algorithm, probability principal component analysis and the like to the noise reduction and feature extraction of a vibration signal of the slewing bearing of the experiment table, and simultaneously applies machine learning methods such as a support vector machine, a BP neural network and the like to the fault diagnosis and service life prediction of the slewing bearing of the experiment table. However, the data is from the laboratory bench, and the fault is formed by artificial processing, so that the fault is still different from the actual fault on the site. Therefore, further research on how to acquire and post-process the support vibration signal during actual use to obtain the fault diagnosis result is still needed.
Disclosure of Invention
In view of the above, the present invention provides a method, an apparatus and a storage medium for diagnosing a fault of a slewing bearing, which are reliable, accurate and fast in response.
In order to achieve the technical purpose, the technical scheme adopted by the invention is as follows:
a slewing bearing fault diagnosis method comprises the following steps:
s1, acquiring a vibration signal of the slewing bearing, and setting the vibration signal as an original signal;
s2, processing the original signal by an EEMD (Ensemble Empirical Mode Decomposition), then obtaining an intrinsic Mode component IMF of the vibration signal by Decomposition, and then carrying out preferential processing on the intrinsic Mode component IMF to obtain an optimal component;
s3, carrying out MCKD parameter optimization processing on the optimal component by using a GWO algorithm and taking the correlation kurtosis as a fitness function to obtain an optimal parameter combination, wherein the optimal parameter combination comprises an input parameter, a period T and a filter length L of the MCKD algorithm;
s4, substituting the optimal parameter combination obtained by the optimization processing of the GWO algorithm into a maximum correlation kurtosis deconvolution algorithm MCKD to analyze the optimal component to obtain a noise reduction signal;
and S5, carrying out envelope spectrum analysis on the noise reduction signal, and comparing and analyzing the fault characteristic frequency found in the envelope spectrum with the theoretical fault characteristic frequency to obtain a diagnosis result.
As a possible implementation, further, the method for processing and decomposing the original signal by the EEMD method is:
and adding Gaussian white noise with equal amplitude for multiple times into the original signal according to the preset standard deviation and the preset times, performing EMD decomposition, and then averaging IMF components subjected to EMD decomposition for multiple times to eliminate the Gaussian white noise added for multiple times.
As a preferred implementation option, preferably, the specific method for processing and decomposing the original signal by the EEMD method includes the following steps:
s21, setting the times of adding the white Gaussian noise and the amplitude of the white Gaussian noise;
s22, adding a uniform white Gaussian noise n to the original signal x (t)i(t) generating a new signal, the mathematical model of which is as follows:
xi(t)＝x(t)+ni(t)
wherein n isi(t) is the i-th addition of white Gaussian noise, xi(t) is the noisy signal for the ith trial, i ═ 1,2, …, N;
s23, for signal x with noisei(t) performing EMD decomposition to obtain a plurality of IMFs and a remainder:
wherein, ci,j(t) is the jth IMF, r after the noisy signal decomposition of the ith experimenti(t) is a residual function;
s24, repeating the step S22 and the step S23 for N times, and averaging the obtained IMFs and the rest items to obtain the final IMF and the rest items:
wherein, cj(t) is the jth IMF after EMD decomposition, and r (t) is the remainder after EMD decomposition.
As a preferred implementation option, preferably, the method for performing the preferential processing on the IMF is: selecting at least one intrinsic mode component IMF as an optimal component by taking a correlation coefficient, an arrangement entropy, a variance contribution rate or a kurtosis as an index, for example, taking the IMF with the largest index value or taking the first 3 IMFs from big to small to reconstruct and synthesize the optimal component, or carrying out normalization processing on the index value, then taking the component with the value larger than 0.5 to reconstruct and the like.
As a preferred implementation choice, in step S1, the number of gaussian white noise additions is preferably 50 or 100, and the amplitude of gaussian white noise is preferably 0.01 to 0.4.
As a preferred implementation choice, preferably, the method for performing MCKD parameter optimization on the optimal component by using the GWO algorithm and using the correlation kurtosis as a fitness function includes:
s31, initializing GWO algorithm to generate a gray wolf group position;
s32, guiding the grey wolf positions in the grey wolf group positions into an MCKD algorithm to calculate the fitness function values CK of the grey wolfs;
s33, carrying out comparative evaluation on the grey wolf fitness function values CK;
s34, updating the gray wolf position according to the evaluation result;
and S35, judging the iterative computation times, returning the updated grey wolf position parameters to the step S32 to be processed step by step when the current computed iterative times are less than the preset maximum iterative times, and outputting the updated grey wolf position parameters to obtain the optimal parameter combination when the current computed iterative times are more than the preset maximum iterative times.
As a possible implementation manner, further, the sampling time length of the vibration signal is longer than the time consumed by one rotation of the slewing bearing, and the sampling frequency of each sampling is 400 Hz.
Based on the diagnosis method, the invention also provides a slewing bearing fault diagnosis device, which comprises:
the acceleration sensors are arranged on the vibration signal detectable area of the slewing bearing and are used for collecting vibration signals generated when the slewing bearing works;
the data processing unit is used for acquiring the vibration signal, processing the vibration signal by an EEMD method and outputting an optimal component after preferential processing;
the parameter optimization unit is used for obtaining the optimal parameter combination of the maximum correlation kurtosis deconvolution algorithm MCKD through GWO algorithm optimization processing, and substituting the optimal parameter combination into the MCKD to analyze the optimal component to obtain a noise reduction signal;
and the diagnosis analysis unit is used for carrying out envelope spectrum analysis on the noise reduction signal and outputting a diagnosis result according to the envelope spectrum analysis.
Based on the diagnosis device, the invention also provides a fault diagnosis device for the slewing bearing of the crane, which comprises the fault diagnosis device for the slewing bearing;
the acceleration sensors are arranged on the leeward side of the portal crane and close to the position of the slewing bearing;
in addition, the slewing bearing is uniformly covered with acceleration sensors in the axial direction and the radial direction.
Based on the above diagnosis method, the present invention further provides a storage medium, where at least one instruction, at least one program, a code set, or an instruction set is stored in the storage medium, and the at least one instruction, at least one program, a code set, or an instruction set is loaded by a processor and executed to implement the slewing bearing fault diagnosis method.
By adopting the technical scheme, compared with the prior art, the invention has the beneficial effects that: the scheme of the invention combines EEMD and MCKD methods for diagnosing the fault of the slewing bearing of the gantry crane. In order to realize the self-adaptive selection of the MCKD parameters, a grayish wolf optimization algorithm is adopted, and the optimal parameter combination is globally optimized by taking the correlation kurtosis as a fitness function; by the method, the extraction of the fault characteristic frequency of the slewing bearing of the portal crane is realized; based on the scheme, under the condition of quick response, the diagnosis and result output of the abnormal vibration in the slewing bearing work can be realized with high accuracy, and reliable and valuable feedback is provided for maintenance personnel.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a diagnostic method of the present invention;
FIG. 2 is a schematic flow diagram of the MCKD of the diagnostic method of the present invention;
FIG. 3 is a time domain diagram and a frequency domain diagram of a slewing bearing in the practice of the present invention;
FIG. 4 is a diagram illustrating the kurtosis of IMF components according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of MCKD parameter optimization;
FIG. 6 is a schematic diagram of an envelope spectrum of an optimal IMF component after MCKD noise reduction;
fig. 7 is a schematic connection diagram of the slewing bearing fault diagnosis device of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be noted that the following examples are only illustrative of the present invention, and do not limit the scope of the present invention. Similarly, the following examples are only some but not all examples of the present invention, and all other examples obtained by those skilled in the art without any inventive work are within the scope of the present invention.
Referring to fig. 1, the slewing bearing fault diagnosis method of the embodiment adopts a method of combining EEMD and GWO-MCKD for extracting fault characteristics of the slewing bearing of the low-speed heavy-load gantry crane, and includes the following specific steps:
(1) parameters of the EEMD algorithm are determined. There are two parameters in EEMD that need to be set manually: standard deviation of gaussian white noise and the number of times noise is added. The standard deviation is determined according to the noise in the signal, and is generally 0.01-0.4, and 0.2 is selected in the embodiment; the number of times of adding noise is usually 50 or 100, and 50 is selected in the embodiment;
(2) performing EMD on the original vibration signal, and selecting the IMF with the maximum kurtosis as the optimal component;
(3) the MCKD best parameter optimization is performed on the best component using the correlation kurtosis as a fitness function using the GWO algorithm. The optimization range for the filter length L is set to [100,700%]. Deconvolution period T ═ fsF, wherein fsIs the sampling frequency and f is the fault signature frequency. Because the fault characteristic frequency is unknown, the theoretical fault characteristic frequency is calculated according to the parameters of the slewing bearing, and the maximum value and the minimum value are substituted into a formula to obtain the range of the deconvolution period T. In addition, due to the fact that errors exist between the theoretical fault characteristic frequency and the actual fault characteristic frequency, the range of T obtained through calculation is properly enlarged;
(4) substituting the optimum parameter combination obtained by GWO optimization into MCKD to analyze the optimum component and make an envelope spectrum;
(5) and comparing and analyzing the fault characteristic frequency found in the envelope spectrum with the theoretical fault characteristic frequency to obtain a diagnosis result.
The EEMD principle and the algorithm flow are as follows:
the EEMD method is characterized in that Gaussian white noise with equal amplitude is added into an original signal for multiple times, the uniform frequency characteristic of the Gaussian white noise is utilized, IMF components after multiple EMD decomposition are averaged, the multiple-time-added Gaussian white noise is eliminated, and meanwhile modal aliasing caused by uneven extreme value distribution is effectively inhibited. The steps of EMD decomposition are as follows:
(1) setting the number of times of adding Gaussian white noise, namely the overall average number of times N;
(2) adding a uniform white Gaussian noise n to the original signal x (t)i(t), generating a new signal:
xi(t)＝x(t)+ni(t)
wherein n isi(t) is the i-th addition of white Gaussian noise, xi(t) is the noisy signal for the ith trial, i ═ 1,2, …, N;
(3) for signal x with noisei(t) performing EMD decomposition to obtain a plurality of IMFs and a remainder:
wherein, ci,j(t) is the jth IMF, r after the noisy signal decomposition of the ith experimenti(t) is a residual function;
(4) repeating the step (2) and the step (3) for N times, and averaging the obtained IMFs and the remainder to obtain the final IMF and the remainder:
wherein, cj(t) is the jth IMF after EMD decomposition, and r (t) is the remainder after EMD decomposition.
In addition, on the basis of fig. 1, referring to fig. 2 for emphasis, the MCKD principle and algorithm flow are as follows:
assuming that a periodic signal generated when the slewing bearing fails is y (t), a transmission path attenuation response is h (t), and doped environmental noise is e (t) when the signal is acquired, the actually measured signal x (t) is:
x(t)＝h(t)*y(t)+e(t)
the essence of the MCKD algorithm is to find the optimal filter f such that the periodic signal y (t) is recovered as much as possible from the measured signal x (t). The algorithm highlights periodic signals submerged in environmental noise by the maximum index of the correlation kurtosis. The correlation kurtosis is calculated as follows:
wherein T is the period of the impulse signal, and M is the shift number. The value range of M is generally 1-7, and if the value is more than 7, the precision is reduced. The parameter M is 5 in this embodiment.
If the associated kurtosis is the greatest, it is equivalent to:
where L is the filter length.
The final calculated derivation can yield f as:
wherein r is 0, T,2T, …, mT
GWO the algorithm principle is as follows:
the gray wolf optimization algorithm is a group of intelligent optimization algorithms developed by Mirjalli et al inspired by the wild predation activities of gray wolfs. The gray wolf is a living animal and strictly adheres to pyramid level relation. The head wolf at the top layer is alpha wolf which is responsible for making decisions on appetite acquisition and perching positions and has leadership; the second layer is a beta wolf obeying to the alpha wolf and is also the best candidate after the alpha wolf is removed; the third layer is delta wolf obeying to alpha and beta wolf; the last layer is omega wolf, which is the basis of the whole wolf group, and the wolfs of the first three layers also need to be taken. The main steps of the gray wolf hunting are three stages: trace, wrap, attack.
The mathematical model of the gray wolf optimization algorithm is represented as follows:
D＝C·Xp(t)-X(t)
X(t+1)＝Xp(t)-A·D
wherein D is the distance between the wolf cluster and the prey, t is the current iteration number, X is the position of the wolf cluster, A, C are the cooperative coefficientsThe vector, whose expression is: a is 2a r1-a，C＝2r2. In the formula r1,r2∈[0,1]And a is reduced from 2 to 0 throughout the iteration.
The course of searching for the prey is mainly done by the guidance of alpha, beta, delta wolf. Therefore, in the iterative process, the currently optimal three grayfans are kept as α, β, δ fans, and the positions of the fan groups are updated according to their positions, which can be expressed as:
Dα＝C1·Xα-X
Dβ＝C2·Xβ-X
Dδ＝C3·Xδ-X
X1＝Xα-A1·Dα
X2＝Xβ-A2·Dβ
X3＝Xδ-A3·Dδ
wherein D isα、Dβ、DδIs the distance between alpha, beta, delta wolf and prey, Xα、Xβ、XδIs the position vector of alpha, beta, delta wolf.
As an implementation example, based on the analysis data collection of the slewing bearing in a certain port portal crane, the following demonstration is carried out, wherein the slewing bearing structure is of a three-row ball type, and the operation mode is that an inner ring is fixed and an outer ring rotates. And an acceleration sensor is selected for data acquisition. Since the arrangement of the sensors directly affects the quality of the acquired data, the positions where the sensors are placed are analyzed first. In order to minimize noise interference and signal transmission quality, the acceleration sensor is placed on the leeward side of the gantry crane and close to the slewing bearing. In addition, the slewing bearing is mainly acted by axial force, radial force and overturning moment, and sensors are arranged in the axial direction and the radial direction.
And selecting the conditions of stable load and speed to acquire data according to the actual working environment on site. The rotating speed of the slewing bearing is 1.22r/min during collection, and the sampling frequency is 400 Hz. Meanwhile, in order to avoid missing fault information, the sampling time is longer than the time of one rotation of the slewing bearing.
The three-row ball type slewing bearing can be simplified into a common three-row rolling bearing. Therefore, the theoretical failure characteristic frequency calculation formula can use the calculation formula of the rolling bearing. The specific formula is as follows:
wherein f isi、fo、frRespectively is a theoretical inner ring fault characteristic frequency, a theoretical outer ring fault characteristic frequency and a theoretical rolling body fault characteristic frequency, D is the pitch diameter of the slewing bearing, D is the diameter of the rolling body, z is the number of the rolling bodies, alpha is a contact angle, fsIs the frequency conversion.
The theoretical fault characteristic frequency of the slewing bearing can be calculated according to the formulas 13 to 15, and is shown in the table 1.
TABLE 1 theoretical failure characteristic frequency
In order to avoid the loss of fault information, data of two revolutions (98s) of the slewing bearing are selected for analysis. Fig. 3 is a time domain and frequency spectrum plot of this data. The apparent impact components are clearly observed in the time domain plot, and many prominent frequencies are observed in the spectrogram, but the characteristic frequencies of the fault cannot be directly distinguished from the prominent frequencies. Therefore, the original signal needs to be processed and further analyzed.
The field signal is analyzed according to the method flow shown in fig. 1.
First, the original signal is decomposed by using the EEMD algorithm to obtain 14 IMFs, and the kurtosis value of each IMF is calculated as shown in fig. 4. As can be seen from fig. 4, the kurtosis value of IMF2 is the largest, which is selected as the optimal component of the analysis; secondly, according to the formula of the deconvolution period, the maximum and minimum theoretical fault characteristic frequencies are selected from the table 1 and substituted, so that the theoretical optimization range of the deconvolution period T is obtained as [96,492], and the range is properly expanded as [90,500 ]. The parameter optimization process for MCKD is shown in FIG. 5. As can be seen from the figure, the correlation kurtosis reaches the maximum at the 7 th generation of the algorithm, and the corresponding optimal combination of L and T is [689,97 ]; finally, setting the filter length L of the MCKD to 689 and the deconvolution period T to 97, and obtaining an envelope spectrum of the optimal component after the MCKD processing, as shown in fig. 6. It can be observed from the figure that there is a fault signature frequency of 4.2Hz and its multiples in the envelope spectrum. Compared with the theoretical fault characteristic frequency in the table 1, the fault characteristic frequency is closest to the theoretical inner ring fault characteristic frequency 4.124Hz of the middle-row slewing bearing. The characteristic frequency has a certain error due to the manufacturing and installation error of the slewing bearing and the abrasion and relative sliding during use. Therefore, the inner ring of the middle row slewing bearing can be judged to be seriously damaged. The process and the result of the whole method show that the fault characteristic frequency is accurately extracted, thereby verifying the effectiveness of the method.
As shown in fig. 7, based on the diagnosis method, the present embodiment further provides a slewing bearing fault diagnosis device, including:
the acceleration sensors 1 are arranged on the vibration signal detectable area of the slewing bearing and are used for collecting vibration signals generated when the slewing bearing works;
the data processing unit 2 is used for acquiring the vibration signal, processing the vibration signal by an EEMD method and outputting an optimal component after preferential processing;
the parameter optimization unit 3 is used for obtaining an optimal parameter combination of the maximum correlation kurtosis deconvolution algorithm MCKD through GWO algorithm optimization processing, and substituting the optimal parameter combination into the MCKD to analyze an optimal component to obtain a noise reduction signal;
and the diagnosis analysis unit 4 is used for carrying out envelope spectrum analysis on the noise reduction signal and outputting a diagnosis result according to the envelope spectrum analysis.
Based on the diagnosis device, the embodiment also provides a fault diagnosis device for the slewing bearing of the crane, which comprises the fault diagnosis device for the slewing bearing;
the acceleration sensors are arranged on the leeward side of the portal crane and close to the position of the slewing bearing;
in addition, the slewing bearing is uniformly covered with acceleration sensors in the axial direction and the radial direction.
Each functional unit in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be substantially or partially implemented in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, a network device, or the like) or a processor (processor) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only a part of the embodiments of the present invention, and not intended to limit the scope of the present invention, and all equivalent devices or equivalent processes performed by the present invention through the contents of the specification and the drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A method of diagnosing a fault in a slewing bearing, comprising:
acquiring a vibration signal of the slewing bearing, and setting the vibration signal as an original signal;
processing an original signal by an EEMD method, decomposing to obtain an intrinsic mode component IMF of a vibration signal, and then performing preferential processing on the intrinsic mode component IMF to obtain an optimal component;
carrying out MCKD parameter optimization processing on the optimal component by using GWO algorithm and taking the correlation kurtosis as a fitness function to obtain an optimal parameter combination;
substituting the optimal parameter combination obtained by the optimization processing of the GWO algorithm into a maximum correlation kurtosis deconvolution algorithm MCKD to analyze the optimal component to obtain a noise reduction signal;
and carrying out envelope spectrum analysis on the noise reduction signal, and comparing and analyzing the fault characteristic frequency found in the envelope spectrum with the theoretical fault characteristic frequency to obtain a diagnosis result.
2. The method of diagnosing a failure in a slewing bearing according to claim 1, wherein the method of processing and decomposing the original signal by the EEMD method comprises:
and adding Gaussian white noise with equal amplitude for multiple times into the original signal according to the preset standard deviation and the preset times, performing EMD decomposition, and then averaging IMF components subjected to EMD decomposition for multiple times to eliminate the Gaussian white noise added for multiple times.
3. The method of diagnosing a failure in a slewing bearing according to claim 2, wherein the method of processing and decomposing the original signal by the EEMD method comprises the steps of:
s21, setting the times of adding the white Gaussian noise and the amplitude of the white Gaussian noise;
s22, adding a group of uniform Gauss to the original signal x (t)White noise ni(t) generating a new signal, the mathematical model of which is as follows:
xi(t)＝x(t)+ni(t)
wherein n isi(t) is the i-th addition of white Gaussian noise, xi(t) is the noisy signal for the ith trial, i ═ 1,2, …, N;
s23, for signal x with noisei(t) performing EMD decomposition to obtain a plurality of IMFs and a remainder:
wherein, ci,j(t) is the jth IMF, r after the noisy signal decomposition of the ith experimenti(t) is a residual function;
s24, repeating the step S22 and the step S23 for N times, and averaging the obtained IMFs and the rest items to obtain the final IMF and the rest items:
wherein, cj(t) is the jth IMF after EMD decomposition, and r (t) is the remainder after EMD decomposition.
4. The method for diagnosing a fault of a slewing bearing according to claim 3, wherein the IMF is preferentially processed by: and selecting one intrinsic mode component IMF or a plurality of IMFs to reconstruct as an optimal component by taking the correlation coefficient, the permutation entropy, the variance contribution rate or the kurtosis as indexes.
5. The method of diagnosing a failure in a slewing bearing according to claim 3, wherein in step S21, the number of times of adding white Gaussian noise is 50 or 100, and the amplitude of the white Gaussian noise is 0.01 to 0.4.
6. The method of slewing bearing fault diagnosis as claimed in claim 1, wherein the method of performing MCKD parameter optimization on the optimal component using the GWO algorithm with the associated kurtosis as a fitness function comprises:
s31, initializing GWO algorithm to generate a gray wolf group position;
s32, guiding the grey wolf positions in the grey wolf group positions into an MCKD algorithm to calculate the fitness function values CK of the grey wolfs;
s33, carrying out comparative evaluation on the grey wolf fitness function values CK;
s34, updating the gray wolf position according to the evaluation result;
and S35, judging the iterative computation times, returning the updated grey wolf position parameters to the step S32 to be processed step by step when the current computed iterative times are less than the preset maximum iterative times, and outputting the updated grey wolf position parameters to obtain the optimal parameter combination when the current computed iterative times are more than the preset maximum iterative times.
7. The method for diagnosing a fault of a slewing bearing according to claim 1, wherein the sampling time for acquiring the vibration signal is longer than the time consumed by one revolution of the slewing bearing, and the sampling frequency of each sampling is 400 Hz.
8. A slewing bearing fault diagnosis device, characterized by comprising:
the acceleration sensors are arranged on the vibration signal detectable area of the slewing bearing and are used for collecting vibration signals generated when the slewing bearing works;
the data processing unit is used for acquiring the vibration signal, processing the vibration signal by an EEMD method and outputting an optimal component after preferential processing;
the parameter optimization unit is used for obtaining the optimal parameter combination of the maximum correlation kurtosis deconvolution algorithm MCKD through GWO algorithm optimization processing, and substituting the optimal parameter combination into the MCKD to analyze the optimal component to obtain a noise reduction signal;
and the diagnosis analysis unit is used for carrying out envelope spectrum analysis on the noise reduction signal and outputting a diagnosis result according to the envelope spectrum analysis.
9. A crane slewing bearing fault diagnosis device, characterized in that it comprises a slewing bearing fault diagnosis device according to claim 8;
the acceleration sensors are arranged on the leeward side of the portal crane and close to the position of the slewing bearing;
in addition, the slewing bearing is uniformly covered with acceleration sensors in the axial direction and the radial direction.
10. A storage medium, characterized by: the storage medium stores at least one instruction, at least one program, a set of codes, or a set of instructions, which is loaded and executed by a processor to implement the method of diagnosing a fault in a slewing bearing according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110130106.3A CN112945557B (en) | 2021-01-29 | 2021-01-29 | Slewing bearing fault diagnosis method, slewing bearing fault diagnosis device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110130106.3A CN112945557B (en) | 2021-01-29 | 2021-01-29 | Slewing bearing fault diagnosis method, slewing bearing fault diagnosis device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112945557A true CN112945557A (en) | 2021-06-11 |
| CN112945557B CN112945557B (en) | 2024-01-12 |
Family
ID=76240196
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110130106.3A Active CN112945557B (en) | 2021-01-29 | 2021-01-29 | Slewing bearing fault diagnosis method, slewing bearing fault diagnosis device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112945557B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113361205A (en) * | 2021-06-23 | 2021-09-07 | 合肥工业大学 | Bearing vibration signal fault diagnosis optimization design method based on genetic algorithm and CEEMD |
| CN113780056A (en) * | 2021-07-22 | 2021-12-10 | 山东科技大学 | Rolling bearing fault diagnosis method based on vibration signal |
| CN114112385A (en) * | 2021-10-29 | 2022-03-01 | 南京理工大学 | Method for measuring natural frequency of ball screw pair feeding system in non-stop state |
| CN114323649A (en) * | 2021-12-31 | 2022-04-12 | 福州大学 | Slewing bearing fault diagnosis method based on CEEMDAN and PSO-MOMEDA |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103278711A (en) * | 2013-04-28 | 2013-09-04 | 福州大学 | Method for early detecting and identifying short-circuit faults and series arc faults |
| CN107084854A (en) * | 2017-04-17 | 2017-08-22 | 四川大学 | Self-adapting random resonant Incipient Fault Diagnosis method based on grey wolf optimized algorithm |
| CN108426715A (en) * | 2018-06-13 | 2018-08-21 | 福州大学 | Rolling bearing Weak fault diagnostic method based on PSO-VMD-MCKD |
| US20190025813A1 (en) * | 2016-05-09 | 2019-01-24 | Strong Force Iot Portfolio 2016, Llc | Methods and systems for intelligent collection and analysis of vehicle data |
| CN109632311A (en) * | 2019-01-21 | 2019-04-16 | 北京化工大学 | A kind of adaptive acoustical signal Method for Bearing Fault Diagnosis |
| WO2019153388A1 (en) * | 2018-02-12 | 2019-08-15 | 大连理工大学 | Power spectral entropy random forest-based aeroengine rolling bearing fault diagnosis method |
| CN111024433A (en) * | 2019-12-30 | 2020-04-17 | 辽宁大学 | Industrial equipment health state detection method for optimizing support vector machine by improving wolf algorithm |
| CN111428418A (en) * | 2020-02-28 | 2020-07-17 | 贵州大学 | Bearing fault diagnosis method and device, computer equipment and storage medium |
| CN111898556A (en) * | 2020-08-01 | 2020-11-06 | 华东交通大学 | Method for extracting transient impact characteristics of bearings based on CK index consistency |
-
2021
- 2021-01-29 CN CN202110130106.3A patent/CN112945557B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103278711A (en) * | 2013-04-28 | 2013-09-04 | 福州大学 | Method for early detecting and identifying short-circuit faults and series arc faults |
| US20190025813A1 (en) * | 2016-05-09 | 2019-01-24 | Strong Force Iot Portfolio 2016, Llc | Methods and systems for intelligent collection and analysis of vehicle data |
| CN107084854A (en) * | 2017-04-17 | 2017-08-22 | 四川大学 | Self-adapting random resonant Incipient Fault Diagnosis method based on grey wolf optimized algorithm |
| WO2019153388A1 (en) * | 2018-02-12 | 2019-08-15 | 大连理工大学 | Power spectral entropy random forest-based aeroengine rolling bearing fault diagnosis method |
| CN108426715A (en) * | 2018-06-13 | 2018-08-21 | 福州大学 | Rolling bearing Weak fault diagnostic method based on PSO-VMD-MCKD |
| CN109632311A (en) * | 2019-01-21 | 2019-04-16 | 北京化工大学 | A kind of adaptive acoustical signal Method for Bearing Fault Diagnosis |
| CN111024433A (en) * | 2019-12-30 | 2020-04-17 | 辽宁大学 | Industrial equipment health state detection method for optimizing support vector machine by improving wolf algorithm |
| CN111428418A (en) * | 2020-02-28 | 2020-07-17 | 贵州大学 | Bearing fault diagnosis method and device, computer equipment and storage medium |
| CN111898556A (en) * | 2020-08-01 | 2020-11-06 | 华东交通大学 | Method for extracting transient impact characteristics of bearings based on CK index consistency |
Non-Patent Citations (6)
| Title |
|---|
| 张丽坪等: "基于相关峭度及自适应变分模态分解的滚动轴承故障诊断", 《轴承》 * |
| 张丽坪等: "基于相关峭度及自适应变分模态分解的滚动轴承故障诊断", 《轴承》, no. 10, 5 October 2020 (2020-10-05), pages 64 - 69 * |
| 张洪梅等: "自适应MCKD和CEEMDAN的滚动轴承微弱故障特征提取", 《电子测量与仪器学报》 * |
| 张洪梅等: "自适应MCKD和CEEMDAN的滚动轴承微弱故障特征提取", 《电子测量与仪器学报》, no. 04, 15 April 2019 (2019-04-15), pages 79 - 86 * |
| 杨伟才等: "基于EEMD降噪和滑动峰态解调的滚动轴承故障诊断方法", 《煤矿机械》 * |
| 杨伟才等: "基于EEMD降噪和滑动峰态解调的滚动轴承故障诊断方法", 《煤矿机械》, no. 01, 15 January 2018 (2018-01-15), pages 127 - 130 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113361205A (en) * | 2021-06-23 | 2021-09-07 | 合肥工业大学 | Bearing vibration signal fault diagnosis optimization design method based on genetic algorithm and CEEMD |
| CN113780056A (en) * | 2021-07-22 | 2021-12-10 | 山东科技大学 | Rolling bearing fault diagnosis method based on vibration signal |
| CN113780056B (en) * | 2021-07-22 | 2024-01-05 | 山东科技大学 | Rolling bearing fault diagnosis method based on vibration signals |
| CN114112385A (en) * | 2021-10-29 | 2022-03-01 | 南京理工大学 | Method for measuring natural frequency of ball screw pair feeding system in non-stop state |
| CN114323649A (en) * | 2021-12-31 | 2022-04-12 | 福州大学 | Slewing bearing fault diagnosis method based on CEEMDAN and PSO-MOMEDA |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112945557B (en) | 2024-01-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112945557A (en) | Slewing bearing fault diagnosis method and device and storage medium | |
| CN110866365B (en) | Mechanical equipment intelligent fault diagnosis method based on partial migration convolutional network | |
| CN113780056B (en) | Rolling bearing fault diagnosis method based on vibration signals | |
| CN104713728B (en) | On-line large slewing bearing residual life prediction method based on multi-dimensional data drive | |
| CN112183581A (en) | Semi-supervised mechanical fault diagnosis method based on self-adaptive migration neural network | |
| CN108171263A (en) | Based on the Fault Diagnosis of Roller Bearings for improving variation mode decomposition and extreme learning machine | |
| CN112418277A (en) | Method, system, medium, and apparatus for predicting remaining life of rotating machine component | |
| CN109556863B (en) | MSPAO-VMD-based large turntable bearing weak vibration signal acquisition and processing method | |
| CN112926505B (en) | Rotating machine health index construction method based on DTC-VAE neural network | |
| CN112633098B (en) | Rotary machine fault diagnosis method, system and storage medium | |
| Omoregbee et al. | Fault detection in roller bearing operating at low speed and varying loads using Bayesian robust new hidden Markov model | |
| CN115017945A (en) | Mechanical fault diagnosis method and system based on enhanced convolutional neural network | |
| CN111060316A (en) | Rolling bearing state monitoring method and system based on convolutional neural network model | |
| CN113776837B (en) | Rolling bearing fault diagnosis method based on CEEMDAN and GWO-NLM | |
| CN112461543A (en) | Rotary machine fault diagnosis method based on multi-classification support vector data description | |
| CN114298110A (en) | Rolling bearing fault diagnosis method and system based on interpretable 1DCNN model | |
| CN114896559A (en) | Self-adaptive noise complete set empirical mode decomposition method based on permutation entropy | |
| CN114264478A (en) | Diesel engine crankshaft bearing wear degree prediction method and system | |
| CN113916535A (en) | Bearing diagnosis method, system, equipment and medium based on time frequency and CNN | |
| Jianhui et al. | New fault diagnosis method for rolling bearing based on PCA | |
| CN113269221A (en) | Rolling bearing fault diagnosis method based on multi-scale convolution neural network | |
| CN116296377A (en) | Industrial robot RV reducer fault prediction method and device | |
| CN112068519A (en) | Quality control method and device for product assembly process | |
| CN114964783B (en) | Gearbox fault detection model based on VMD-SSA-LSSVM | |
| Yadav et al. | Variable dropout one-dimensional cnn for vibration-based shaft unbalance detection in industrial machinery |
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 |