CN107271184A - The kernel regression decomposition method and system of a kind of rolling bearing fault diagnosis - Google Patents
The kernel regression decomposition method and system of a kind of rolling bearing fault diagnosis Download PDFInfo
- Publication number
- CN107271184A CN107271184A CN201710367859.XA CN201710367859A CN107271184A CN 107271184 A CN107271184 A CN 107271184A CN 201710367859 A CN201710367859 A CN 201710367859A CN 107271184 A CN107271184 A CN 107271184A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- kernel regression
- msup
- formula
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
Abstract
The invention discloses the kernel regression decomposition method and system of a kind of rolling bearing fault diagnosis.The present invention includes motor, V belt translation axle, shaft coupling, test bearing, acceleration transducer, multi-channel data acquisition analyzer, computer；The faulty bearings are connected by motor, V belt translation axle and shaft coupling, acceleration transducer is fixed on test bearing bearing block, acceleration transducer output end is connected with multi-channel data acquisition analyzer, data after analyzer extraction after preservation by being sent to computer, and signal data is analyzed and processed on computers with reference to kernel regression decomposition method, bearing combined failure is identified, the accurate detection to rolling bearing running status is realized.The present invention can effectively gather the fault message of rolling bearing, and good with transmission performance, accuracy rate is done, and speed is fast, simple operation and other advantages, with engineering application value.
Description
Technical field
The present invention relates to mechanical fault diagnosis field, a kind of kernel regression of rolling bearing fault diagnosis point is specifically referred to
Solve method and system.
Background technology
Rolling bearing is wide variety of important mechanical part in all kinds of machines, is also the element for being easiest in machine damage
One of.Rotating machinery is the health status of rolling body bearing of placing one's entire reliance upon, and almost accounts for 40- 50% equipment fault.Axle
The failure held is probably and its serious that may cause the shut-down of whole production line, even result in the injures and deaths of personnel.At this stage
Fault diagnosis to bearing mainly judges the running status of bearing by artificial experience or instrument, but is actually obtaining vibration signal
In often due to working condition is complicated, ambient noise big, equipment is in the reasons such as working condition for a long time, be sometimes difficult to obtain therefore
Hinder the obvious signal of aspect ratio.Therefore, signal denoising has become the committed step of bearing fault signal transacting.Current event
Hinder the limitations such as diagnostic system existence and stability is poor, cumbersome, can not still be widely used popularization.In order to solve to roll
The technical problems such as bearing failure diagnosis difficulty, False Rate height, the research of the fault diagnosis system of bearing also becomes more and more intentional
Justice.
In existing all kinds of bearing failure diagnosis technologies, the analysis of vibration signal is still a kind of main method.Base
Represented in the signal frequency domain of Fourier transformation, disclose the inner link between the function of time and frequency spectrum function, put down in traditional
Extremely important effect has been played in steady signal analysis and processing, but when extracting signal spectrum with the method for Fourier transformation, has been needed
Whole time-domain informations of signal are utilized, this is a kind of integral transformation, lacks time domain positioning function.Wavelet transformation is in recent years
A kind of new transform analysis method, it inherits and developed the thought of short time discrete Fourier transform localization, while overcoming window again
The shortcomings of mouthful size does not change with frequency, wavelet transformation can more preferable observation signal local characteristicses, letter can be observed simultaneously
Number time and frequency information, this is that Fourier transformation does not reach.But suppress noise often in processing using wavelet transformation
Cause oscillation effect during Low SNR signal, while need to manually select suitable small echo when carrying out wavelet transformation, it is so right
The required time is considerably increased in the processing of Practical Project problem.EMD (the experiences that American Engineering academician doctor Huang E proposes
Mode decomposition) a kind of adaptive data processing and method for digging are had proven to, it is especially suitable for non-linear and non-stationary
The processing of time series, and any basic function need not be preset, it is substantially also the tranquilization to data sequence or signal
Processing.The purpose of EMD algorithms is, by the intrinsic mode functions that the bad signal decomposition of performance is one group of better performances, to be decomposited
Each component come contains the local feature signal of the different time scales of original signal.The suitable component of selection, is then carried out
Hilbert transform obtains time-frequency spectrum, obtains the frequency of physical significance.But in the work environment, how correctly to select addition
Noise amplitude still needs to further research, and frequently occurring for modal overlap is also one of EMD major defect, and modal overlap is signal
Caused by interrupting, interruption is a kind of disturbing signal of indefinite form, situation about being frequently encountered in actual treatment.Interruption can be led
Time-frequency distributions are obscured in cause, and then destroy IMF physical significance.
The content of the invention
Technical problem to be solved of the embodiment of the present invention is that there is provided a kind of kernel regression of rolling bearing fault diagnosis point
Method and system are solved, and Treatment Analysis is carried out to vibration signal using kernel regression decomposition method, bearing correlated characteristic is extracted and goes forward side by side
Row identification.
To achieve the above object, the technical scheme is that comprising the following steps：
(1) data acquisition, the vibration signal data of rolling bearing to be measured is gathered using acceleration transducer, and the acceleration is passed
Sensor is arranged on the bearing block of bearing to be measured, and acceleration transducer output end is connected with multi-channel data acquisition analyzer, many
Channel data acquisition and analysis instrument is by the data after extraction by being sent to computer after preservation；
(2) signal transacting is carried out to the real time data measured by acceleration transducer using kernel regression decomposition method, core is returned
Return decomposition method：
(2.1) data for first obtaining multi-channel data acquisition analyzer are carried out as primary signal with gaussian kernel function
The processing of first time kernel regression obtains a remaining variable, and the formula that kernel regression used is decomposed is：
In formula：f∑,1(t) it is primary signal, f∑,1(ti) it is kernel function center, λ1For bandwidth parameter, f∑,2(t) it is
The new signal obtained after kernel regression conversion, K1For gaussian kernel function；
(2.2) and then by the remaining variable of gained and original signal subtract each other that to obtain a new representation in components be first remnants
Component, then to f∑,2(t) kernel regression formula manipulation is carried out again, and new residual components, but all points are continuously available by that analogy
Solution is not endless decomposition, stops decomposing when meeting the standard deviation standard that we give, its formula is:
In formula:T is the length of data, Cj(t) component to be obtained with kernel regression；When meeting the threshold value that is set in above formula,
Kernel regression processing stops carrying out next step analysis；
Then, for without processing it is complete caused by some unnecessary information included in each new residual components
Soft-threshold processing is carried out, the component of gained is further subjected to denoising Processing, its threshold formula is：
θj=MAD (Cj(t))/0.6745 (3)
In formula：MAD is mean absolute deviation；
(2.3) final step of kernel regression is exactly that all obtained residual components are reconstructed, by will be all residual
Remaining component carries out obtaining new signal to carry out data analysis and fine fault diagnosis after adding up；
(3) finally all components are superimposed and obtain carrying out Hilbert envelope analysis of spectrum after new signal, in time-frequency
Rolling bearing fault information is drawn in conversion.
Further set is that the threshold value that the kernel regression processing in the step (2.2) set by formula (2) stops is 0.2.
The present invention also provides a kind of Diagnosing System for Detecting of Antifriction Bearings based on kernel regression decomposition method, includes for pacifying
Fill the power transmission shaft of rolling bearing to be measured, for driving the power unit for being located at drive axis, the bearing for being arranged at bearing to be measured
On seat by gather roller bearing acceleration information to be measured acceleration transducer and multi-channel data acquisition analyzer and based on
Calculation machine, acceleration transducer output end is connected with multi-channel data acquisition analyzer, and multi-channel data acquisition analyzer will be extracted
Data afterwards after preservation by being sent to computer；
Signal is carried out to the real time data measured by acceleration transducer using kernel regression decomposition method in the computer
Processing, kernel regression decomposition method：
The data that multi-channel data acquisition analyzer is obtained first carry out first as primary signal with gaussian kernel function
Secondary kernel regression processing obtains a remaining variable, and the formula that kernel regression used is decomposed is：
In formula：f∑,1(t) it is primary signal, f∑,1(ti) it is kernel function center, λ1For bandwidth parameter, f∑,2(t) it is kernel regression
The new signal obtained after conversion, K1For gaussian kernel function；
Then the remaining variable of gained and original signal subtracted each other to obtain a new representation in components be first residual components,
Then to f∑,2(t) kernel regression formula manipulation is carried out again, is continuously available new residual components by that analogy, but all decomposition is simultaneously
It is not endless decomposition, stops decomposing when meeting the standard deviation standard that we give, its formula is:
In formula:T is the length of data, Cj(t) component to be obtained with kernel regression；When meeting the threshold value that is set in above formula,
Kernel regression processing stops carrying out next step analysis；
Then, for without processing it is complete caused by some unnecessary information included in each new residual components
Soft-threshold processing is carried out, the component of gained is further subjected to denoising Processing, its threshold formula is：
θj=MAD (Cj(t))/0.6745 (3)
In formula：MAD is mean absolute deviation；
The final step of kernel regression is exactly that all obtained residual components are reconstructed, by by all residual components
Carry out obtaining new signal to carry out data analysis and fine fault diagnosis after adding up；
Finally all component superpositions are obtained to carry out Hilbert envelope analysis of spectrum after new signal, in time-frequency convert
In draw rolling bearing fault information.
It is that power unit includes motor further to set, and the output shaft of motor is passed by belt transmission unit and power transmission shaft
Dynamic connection, described power transmission shaft installs rolling bearing to be measured by shaft coupling.
Kernel regression method of the present invention is a kind of new digital signal processing method, will first based on EMD principles
Signal is resolved into several yardsticks with kernel regression technology, and each yardstick includes certain characteristic information again simultaneously, but each yardstick is again
Contain some useless information.Therefore, optimize this algorithm with reference to soft threshold method and standard deviation standard, reached with this
The effect of denoising, finally carries out time-frequency convert, by new signal transacting using Hilbert envelope spectral method by signal
Result is obtained afterwards and judges fault message.
By actual application, the present invention has good effect in processing rolling bearing fault signal, and this method is more
Suitable for the denoising in mechanical system and fault detect.
The beneficial effects of the invention are as follows：The present invention has transmission performance good, and speed is fast, simple to operate, and machine is found in time
The failure of rolling bearing, with good engineer applied effect.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, according to
These accompanying drawings obtain other accompanying drawings and still fall within scope of the invention.
Fig. 1 is kernel regression decomposition method trouble-shooting chart of the invention；
Fig. 2 is kernel regression method route figure of the invention；
Fig. 3 is the result of bearing outer ring fault-signal of the present invention；
Fig. 4 is bearing combined failure signal processing results of the present invention.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, the present invention is made into one below in conjunction with accompanying drawing
It is described in detail on step ground.
Direction and position term that the present invention is previously mentioned, such as " on ", " under ", "front", "rear", "left", "right", " interior ", "
Outside ", " top ", " bottom ", " side " etc., are only the direction or position of refer to the attached drawing.Therefore, the direction used and position term
It is to illustrate and understand the present invention, rather than limiting the scope of the invention.
As shown in Figures 1 to 4, be in the embodiment of the present invention, including motor, V belt translation axle, shaft coupling, faulty bearings plus
Velocity sensor, multi-channel data acquisition analyzer, PC, faulty bearings described in Fig. 1 pass through motor, V belt translation axle and shaft coupling
It is connected with power transmission shaft, acceleration transducer is fixed on test bearing bearing block, acceleration transducer output end and multichannel number
Connected according to acquisition and analysis instrument, analyzer is by the data after extraction by being sent to computer, the present embodiment computer after preservation
Using traditional X86 or X64 PC, PC combination kernel regression decomposition technique is analyzed and processed to signal data, realization pair
The accurate detection of rolling bearing running status.Operating procedure is as follows：
(1) data acquisition unit is selected first, and the present invention uses acceleration transducer gathered data, acceleration transducer peace
On the bearing block of bearing to be measured, sample frequency is generally 25600Hz, real in Fig. 2 according to the specific parameter setting of sensor
Test on platform after data measured result in real-time incoming PC.
(2) secondly the present invention will be believed the real time data measured by acceleration transducer using kernel regression decomposition method
Number processing.Kernel regression decomposition method：
The data that multi-channel data acquisition analyzer is obtained first carry out first as primary signal with gaussian kernel function
Secondary kernel regression processing obtains a remaining variable, and the formula that kernel regression used is decomposed is：
In formula：f∑,1(t) it is primary signal, f∑,1(ti) it is kernel function center, λ1For bandwidth parameter, f∑,2(t) it is kernel regression
The new signal obtained after conversion, K1For gaussian kernel function.
Then the remaining variable of gained and original signal subtracted each other to obtain a new representation in components be first residual components,
Then to f∑,2(t) kernel regression formula manipulation is carried out again, and new residual components are continuously available by that analogy.But all decomposition are simultaneously
It is not endless decomposition, stops decomposing when meeting the standard deviation standard that we give.Its formula is:
In formula:T is the length of data, Cj(t) component to be obtained with kernel regression.When meeting the threshold value that is set in above formula,
When typically we are set to 0.2, kernel regression processing stops carrying out next step analysis.
In addition, it is possible to not handle complete in above-mentioned processing procedure, is wrapped again in each new residual components
Contain some unnecessary information, therefore the present invention uses for reference EMD method progress soft-threshold processing, further enters the component of gained
Row denoising Processing.Its threshold formula is：
θj=MAD (Cj(t))/0.6745 (3)
In formula：MAD is mean absolute deviation.
The final step of kernel regression is exactly that all obtained residual components are reconstructed, by by all residual components
Carry out obtaining new signal to carry out data analysis and fine fault diagnosis after adding up.
(3) finally all components are superimposed and obtain carrying out Hilbert envelope analysis of spectrum after new signal, in time-frequency
Bearing fault information is drawn in conversion.Concrete operations flow is with reference to as shown in Figure 2.
As in Fig. 1, described kernel regression decomposition technique bearing failure diagnosis system, fault message is adopted by acceleration transducer
Collection.
Acceleration transducer is arranged on faulty bearings in such as Fig. 1, is placed on the bearing block of test bearing, is carried out radially
The vibration signal sampling in direction.
Such as multi-channel data acquisition analyzer model AVANT-MI-7016 in Fig. 1, there are 16 passages, each passage tool
Have signal acquisition, extraction, filtering, signal source output function, the system is become relatively reliable on hardware and is simplified perhaps
The intermediate steps of multi signal processing, it is practical and convenient.
Referring to Fig. 1, Fig. 1 is a kind of knot of embodiment of Diagnosing System for Detecting of Antifriction Bearings provided by the present invention
Structure block diagram.
Diagnosing System for Detecting of Antifriction Bearings is used for the phenomenon of the failure for detecting rolling bearing, and fault vibration signal is uploaded to
PC.In specific scheme, motor drives V belt translation, and shaft coupling and power transmission shaft provide rotating speed into the testing stand of simulation, and
By acceleration transducer acquisition parameter signal on faulty bearings, the parameter signal of these collections is carried out by kernel regression method
Processing, so as to determine the situation of equipment, the description to this status of equipment is phenomenon of the failure, the inner ring of such as bearing, outer ring,
The phenomena of the failure such as rolling element.Specific computational methods can be calculated with following equation：
Outer ring failure formula：
Inner ring failure formula：
Rolling element failure formula：
In formula：frFor speed, n is bearing roller number, and φ is radial direction contact angle, and d is that rolling element is averagely straight
Footpath, D is the average diameter of bearing.
Example case 1：
It is the result figure by being obtained handled by the Diagnosing System for Detecting of Antifriction Bearings as shown in Figure 3, according to existing axle
The failure-frequency of this experiment axis bearing outer-ring can be calculated for 91.15Hz by holding information, and failure can be rapidly distinguished from figure
Frequency for 87.5Hz it is close with theoretical value, wherein 29.8Hz for the motor speed, it is possible to determine that this rolling bearing
Fault type is bearing outer ring failure.
Case study on implementation 2：
On the basis of example 1, the implementation case 2 is by the more complicated bearing fault of further processing, by calculating this
Faulty bearings are that combined failure includes bearing inner race, outer ring and rolling element failure in experiment.Pass through the data meter to test bearing
The failure-frequency of inner ring can be obtained for 197.05Hz by calculating, and outer ring failure-frequency is 121.51Hz, and rolling element failure is
79.25Hz.Frequency is 120Hz and outer ring Trouble Match as can be drawn from Figure 4, and 160Hz and two frequencys multiplication of rolling element failure are accorded with
Close, 202.5Hz is substantially conformed to inner ring failure.So as to further illustrate that the present invention has preferable treatment effect, it is worthy to be popularized
Using.
Can be with one of ordinary skill in the art will appreciate that realizing that all or part of step in above-described embodiment method is
The hardware of correlation is instructed to complete by program, described program can be stored in a computer read/write memory medium,
Described storage medium, such as ROM/RAM, disk, CD.
Above disclosure is only preferred embodiment of present invention, can not limit the right model of the present invention with this certainly
Enclose, therefore the equivalent variations made according to the claims in the present invention, still belong to the scope that the present invention is covered.
Claims (4)
1. the kernel regression decomposition method of a kind of rolling bearing fault diagnosis, it is characterised in that comprise the following steps：
(1) data acquisition, the vibration signal data of rolling bearing to be measured, the acceleration transducer are gathered using acceleration transducer
On the bearing block of bearing to be measured, acceleration transducer output end is connected with multi-channel data acquisition analyzer, multichannel
Data collection and analysis instrument is by the data after extraction by being sent to computer after preservation；
(2) signal transacting, kernel regression point are carried out to the real time data measured by acceleration transducer using kernel regression decomposition method
Solution method：
(2.1) data for first obtaining multi-channel data acquisition analyzer carry out first as primary signal with gaussian kernel function
Secondary kernel regression processing obtains a remaining variable, and the formula that kernel regression used is decomposed is：
<mrow>
<msub>
<mi>f</mi>
<mrow>
<mi>&Sigma;</mi>
<mo>,</mo>
<mn>2</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>K</mi>
<mn>1</mn>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mrow>
<mi>&Sigma;</mi>
<mo>,</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>-</mo>
<msub>
<mi>f</mi>
<mrow>
<mi>&Sigma;</mi>
<mo>,</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>(</mo>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msup>
<mi>e</mi>
<mrow>
<mo>-</mo>
<mfrac>
<mn>1</mn>
<mrow>
<mn>2</mn>
<msup>
<msub>
<mi>&lambda;</mi>
<mn>1</mn>
</msub>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mrow>
<mi>&Sigma;</mi>
<mo>,</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>-</mo>
<msub>
<mi>f</mi>
<mrow>
<mi>&Sigma;</mi>
<mo>,</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>(</mo>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msup>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula：f∑,1(t) it is primary signal, f∑,1(ti) it is kernel function center, λ1For bandwidth parameter, f∑,2(t) converted for kernel regression
The new signal obtained afterwards, K1For gaussian kernel function；
(2.2) and then by the remaining variable of gained and original signal subtract each other that to obtain a new representation in components be first remnants points
Amount, then to f∑,2(t) kernel regression formula manipulation is carried out again, and new residual components, but all decomposition are continuously available by that analogy
It is not endless decomposition, stops decomposing when meeting the standard deviation standard that we give, its formula is:
<mrow>
<mi>S</mi>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>t</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<mi>T</mi>
</munderover>
<mo>&lsqb;</mo>
<mfrac>
<mrow>
<mo>|</mo>
<msub>
<mi>C</mi>
<mrow>
<mi>j</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msub>
<mi>C</mi>
<mi>j</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
</mrow>
<mrow>
<msub>
<msup>
<mi>C</mi>
<mn>2</mn>
</msup>
<mrow>
<mi>j</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>&rsqb;</mo>
<mo>&le;</mo>
<mn>0.2</mn>
<mo>-</mo>
<mn>0.3</mn>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula:T is the length of data, Cj(t) component to be obtained with kernel regression；When meeting the threshold value that is set in above formula, kernel regression
Processing stops carrying out next step analysis；
Then, for without processing it is complete caused by included in each new residual components some unnecessary information carry out
Soft-threshold processing, further carries out denoising Processing by the component of gained, and its threshold formula is：
θj=MAD (Cj(t))/0.6745 (3)
In formula：MAD is mean absolute deviation；
(2.3) final step of kernel regression is exactly that all obtained residual components are reconstructed, by by all remnants points
Amount carries out obtaining new signal to carry out data analysis and fine fault diagnosis after adding up；
(3) finally all components are superimposed and obtain carrying out Hilbert envelope analysis of spectrum after new signal, in time-frequency convert
In draw rolling bearing fault information.
2. a kind of kernel regression decomposition method of rolling bearing fault diagnosis according to claim 1, it is characterised in that：It is described
The threshold value that kernel regression processing in step (2.2) set by formula (2) stops is 0.2.
3. a kind of Diagnosing System for Detecting of Antifriction Bearings based on kernel regression decomposition method, it is characterised in that：Include for installing
The power transmission shaft of rolling bearing to be measured, for driving power unit positioned at drive axis, being arranged at the bearing block of bearing to be measured
Upper acceleration transducer and multi-channel data acquisition analyzer and the calculating for being used to gather roller bearing acceleration information to be measured
Machine, acceleration transducer output end is connected with multi-channel data acquisition analyzer, after multi-channel data acquisition analyzer will be extracted
Data by being sent to computer after preservation；
Signal transacting is carried out to the real time data measured by acceleration transducer using kernel regression decomposition method in the computer,
Kernel regression decomposition method：
The data that multi-channel data acquisition analyzer is obtained first carry out first time core as primary signal with gaussian kernel function
Recurrence processing obtains a remaining variable, and the formula that kernel regression used is decomposed is：
<mrow>
<msub>
<mi>f</mi>
<mrow>
<mi>&Sigma;</mi>
<mo>,</mo>
<mn>2</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>K</mi>
<mn>1</mn>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<msub>
<mi>f</mi>
<mrow>
<mi>&Sigma;</mi>
<mo>,</mo>
<mn>1</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msub>
<mi>f</mi>
<mrow>
<mi>&Sigma;</mi>
<mo>,</mo>
<mn>1</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msup>
<mi>e</mi>
<mrow>
<mo>-</mo>
<mfrac>
<mn>1</mn>
<mrow>
<mn>2</mn>
<msup>
<msub>
<mi>&lambda;</mi>
<mn>1</mn>
</msub>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
<msup>
<mrow>
<mo>(</mo>
<mrow>
<msub>
<mi>f</mi>
<mrow>
<mi>&Sigma;</mi>
<mo>,</mo>
<mn>1</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msub>
<mi>f</mi>
<mrow>
<mi>&Sigma;</mi>
<mo>,</mo>
<mn>1</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>t</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msup>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula：f∑,1(t) it is primary signal, f∑,1(ti) it is kernel function center, λ1For bandwidth parameter, f∑,2(t) converted for kernel regression
The new signal obtained afterwards, K1For gaussian kernel function；
Then the remaining variable of gained and original signal subtracted each other to obtain a new representation in components be first residual components, then
To f∑,2(t) kernel regression formula manipulation is carried out again, and new residual components are continuously available by that analogy, but all decomposition are not
It is endless to decompose, stop decomposing when meeting the standard deviation standard that we give, its formula is:
<mrow>
<mi>S</mi>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>t</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<mi>T</mi>
</munderover>
<mo>&lsqb;</mo>
<mfrac>
<mrow>
<mo>|</mo>
<msub>
<mi>C</mi>
<mrow>
<mi>j</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msub>
<mi>C</mi>
<mi>j</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
</mrow>
<mrow>
<msub>
<msup>
<mi>C</mi>
<mn>2</mn>
</msup>
<mrow>
<mi>j</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>&rsqb;</mo>
<mo>&le;</mo>
<mn>0.2</mn>
<mo>-</mo>
<mn>0.3</mn>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula:T is the length of data, Cj(t) component to be obtained with kernel regression；When meeting the threshold value that is set in above formula, kernel regression
Processing stops carrying out next step analysis；
Then, for without processing it is complete caused by included in each new residual components some unnecessary information carry out
Soft-threshold processing, further carries out denoising Processing by the component of gained, and its threshold formula is：
θj=MAD (Cj(t))/0.6745 (3)
In formula：MAD is mean absolute deviation；
The final step of kernel regression is exactly that all obtained residual components are reconstructed, by the way that all residual components are carried out
Obtain new signal to carry out data analysis and fine fault diagnosis after cumulative；
Finally all component superpositions are obtained to carry out Hilbert envelope analysis of spectrum after new signal, in time-frequency convert
Go out rolling bearing fault information.
4. a kind of Diagnosing System for Detecting of Antifriction Bearings based on kernel regression decomposition method according to claim 3, its feature
It is：Power unit includes motor, and the output shaft of motor is connected by belt transmission unit with transmission shaft driven.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710367859.XA CN107271184B (en) | 2017-05-23 | 2017-05-23 | A kind of the kernel regression decomposition method and system of rolling bearing fault diagnosis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710367859.XA CN107271184B (en) | 2017-05-23 | 2017-05-23 | A kind of the kernel regression decomposition method and system of rolling bearing fault diagnosis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107271184A true CN107271184A (en) | 2017-10-20 |
| CN107271184B CN107271184B (en) | 2019-06-11 |
Family
ID=60064189
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710367859.XA Active CN107271184B (en) | 2017-05-23 | 2017-05-23 | A kind of the kernel regression decomposition method and system of rolling bearing fault diagnosis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107271184B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108694298A (en) * | 2018-07-04 | 2018-10-23 | 温州大学 | A kind of axial plunger pump bearing inner race and outer ring method for diagnosing faults |
| CN109029997A (en) * | 2018-09-13 | 2018-12-18 | 温州大学 | A kind of polynary synchronous compression method of kernel regression remnants decomposition of internal combustion engine Rolling Bearing Status detection |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050261837A1 (en) * | 2004-05-03 | 2005-11-24 | Smartsignal Corporation | Kernel-based system and method for estimation-based equipment condition monitoring |
| CN103345760A (en) * | 2013-07-29 | 2013-10-09 | 常熟理工学院 | Method for automatically generating mark points of object shape template of medical image |
| US8620853B2 (en) * | 2011-07-19 | 2013-12-31 | Smartsignal Corporation | Monitoring method using kernel regression modeling with pattern sequences |
| CN103854262A (en) * | 2014-03-20 | 2014-06-11 | 西安电子科技大学 | Medical image noise reduction method based on structure clustering and sparse dictionary learning |
| CN105678715A (en) * | 2016-02-15 | 2016-06-15 | 江苏大学 | Method for removing image noise based on kernel regression total variation |
-
2017
- 2017-05-23 CN CN201710367859.XA patent/CN107271184B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050261837A1 (en) * | 2004-05-03 | 2005-11-24 | Smartsignal Corporation | Kernel-based system and method for estimation-based equipment condition monitoring |
| US8620853B2 (en) * | 2011-07-19 | 2013-12-31 | Smartsignal Corporation | Monitoring method using kernel regression modeling with pattern sequences |
| CN103345760A (en) * | 2013-07-29 | 2013-10-09 | 常熟理工学院 | Method for automatically generating mark points of object shape template of medical image |
| CN103854262A (en) * | 2014-03-20 | 2014-06-11 | 西安电子科技大学 | Medical image noise reduction method based on structure clustering and sparse dictionary learning |
| CN105678715A (en) * | 2016-02-15 | 2016-06-15 | 江苏大学 | Method for removing image noise based on kernel regression total variation |
Non-Patent Citations (4)
| Title |
|---|
| JIANG LIU: "Fault prediction for power plant equipment based on", 《2015 8TH INTERNATIONAL SYMPOSIUM ON COMPUTATIONAL INTELLIGENCE AND DESIGN》 * |
| JIA-REN CHANG CHIEN: "Design of a Home Care Instrument Based on", 《INDUSTRIAL TECHNOLOGY》 * |
| SHITAO LIU: "Kernel Least Squares Regression for Fault Isolation", 《PROCEEDINGS OF THE 33RD CHINESE CONTROL CONFERENCE》 * |
| 柏叶婷: "统计方法在轴承故障检测中的应用", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108694298A (en) * | 2018-07-04 | 2018-10-23 | 温州大学 | A kind of axial plunger pump bearing inner race and outer ring method for diagnosing faults |
| CN109029997A (en) * | 2018-09-13 | 2018-12-18 | 温州大学 | A kind of polynary synchronous compression method of kernel regression remnants decomposition of internal combustion engine Rolling Bearing Status detection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107271184B (en) | 2019-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cui et al. | Quantitative trend fault diagnosis of a rolling bearing based on Sparsogram and Lempel-Ziv | |
| Li et al. | Early fault feature extraction of rolling bearing based on ICD and tunable Q-factor wavelet transform | |
| Hemmati et al. | Roller bearing acoustic signature extraction by wavelet packet transform, applications in fault detection and size estimation | |
| Cong et al. | Short-time matrix series based singular value decomposition for rolling bearing fault diagnosis | |
| Lei et al. | Application of an improved kurtogram method for fault diagnosis of rolling element bearings | |
| He et al. | Multifractal entropy based adaptive multiwavelet construction and its application for mechanical compound-fault diagnosis | |
| Ocak et al. | HMM-based fault detection and diagnosis scheme for rolling element bearings | |
| Wang et al. | Traversal index enhanced-gram (TIEgram): A novel optimal demodulation frequency band selection method for rolling bearing fault diagnosis under non-stationary operating conditions | |
| CN109932179B (en) | Rolling bearing fault detection method based on DS self-adaptive spectrum reconstruction | |
| Jiang et al. | A time-frequency spectral amplitude modulation method and its applications in rolling bearing fault diagnosis | |
| CN104596766B (en) | Early fault determining method and device for bearing | |
| CN103471848A (en) | Rolling bearing fault feature extraction method based on independent component analysis and cepstrum theory | |
| KR101607047B1 (en) | Signal analysis method and apparatus for fault detection | |
| CN110057583A (en) | A kind of bearing fault recognition methods, device and computer equipment | |
| Klausen et al. | Multi-band identification for enhancing bearing fault detection in variable speed conditions | |
| Song et al. | Self-adaptive multivariate variational mode decomposition and its application for bearing fault diagnosis | |
| Sheng et al. | Applications in bearing fault diagnosis of an improved Kurtogram algorithm based on flexible frequency slice wavelet transform filter bank | |
| CN110160765A (en) | A kind of shock characteristic recognition methods and system based on sound or vibration signal | |
| CN107506710A (en) | A kind of rolling bearing combined failure extracting method | |
| CN112364706A (en) | Small sample bearing fault diagnosis method based on class imbalance | |
| CN108036940A (en) | A kind of Method for Bearing Fault Diagnosis | |
| CN107271184A (en) | The kernel regression decomposition method and system of a kind of rolling bearing fault diagnosis | |
| CN104112003B (en) | The method and system that the performance of game terminal is detected | |
| CN111811819A (en) | Bearing fault diagnosis method and device based on machine learning | |
| CN114689298A (en) | Fault detection method and device for walking part of coal mining machine |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Xiang Jiawei Inventor after: Lou Kai Inventor after: Zhong Yongteng Inventor before: Xiang Jiawei Inventor before: Zhong Yongteng Inventor before: Lou Kai |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20171020 Assignee: Guowang Technology (Zhejiang) Co.,Ltd. Assignor: Wenzhou University Contract record no.: X2023330000103 Denomination of invention: A Kernel Regression Decomposition Method and System for Fault Diagnosis of Rolling Bearings Granted publication date: 20190611 License type: Common License Record date: 20230311 |