CN107917806A - A kind of Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD - Google Patents

A kind of Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD Download PDF

Info

Publication number
CN107917806A
CN107917806A CN201711256531.7A CN201711256531A CN107917806A CN 107917806 A CN107917806 A CN 107917806A CN 201711256531 A CN201711256531 A CN 201711256531A CN 107917806 A CN107917806 A CN 107917806A
Authority
CN
China
Prior art keywords
mrow
msub
mtd
signal
mtr
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201711256531.7A
Other languages
Chinese (zh)
Inventor
孙伟
张先辉
熊天旸
李新民
金小强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Helicopter Research and Development Institute
Original Assignee
China Helicopter Research and Development Institute
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China Helicopter Research and Development Institute filed Critical China Helicopter Research and Development Institute
Priority to CN201711256531.7A priority Critical patent/CN107917806A/en
Publication of CN107917806A publication Critical patent/CN107917806A/en
Pending legal-status Critical Current

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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/02—Preprocessing
    • G06F2218/04—Denoising
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction
    • G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching
    • G06F2218/14—Classification; Matching by matching peak patterns

Landscapes

  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • General Physics & Mathematics (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)

Abstract

It is a kind of Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD the present invention relates to the technical solution adopted by the present invention, belong to fault diagnosis technology field, it pre-processes rolling bearing acceleration signal by using MCKD methods, while noise reduction, strengthen the impact component of signal, and rolling bearing acceleration signal is decomposed by LMD methods, any one complicated non-stationary signal is adaptively resolved into the sum of the PF components of multiple instantaneous frequencys with physical significance, each of which PF components are all to be multiplied to obtain by an envelope signal and a pure FM signal, its instantaneous frequency has physical significance, it is also a process progressively decomposed from high frequency to low frequency that LMD, which is decomposed, at the same time, it is more accurate for fault diagnosis.

Description

A kind of Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD
Technical field
The invention belongs to mechanical fault diagnosis field, more particularly to a kind of rolling bearing early stage event based on MCKD and LMD Hinder diagnostic method.
Background technology
Rolling bearing is widely used in various rotating machineries, its defect and damage will directly affect equipment and stablize fortune OK.The common failure of rolling bearing has inner ring failure, outer ring failure and ball failure etc., these failures can form periodic shock Component.If there is early stage local fault, component unobvious are impacted, and are disturbed be subject to random noise, is easily blanked, causes Failure is not easy to identify.
Local mean value decomposes frequency division when (Local mean decomposition, abbreviation LMD) is a kind of new adaptive Analysis method, since this method can effectively handle non-linear, non-stationary signal, had many scholars to apply this method to rolling in recent years Dynamic bearing fault diagnosis field, such as LMD and enhancing envelope spectrum are combined carry out Analysis of Fault on Roller Bearing, achieve preferably Effect；LMD and order tracking technique analysis are combined, is sampled using order tracking technique and time domain fault vibration signal is transformed into angular domain, LMD is carried out to angular domain signal to decompose, judge the trouble location and type of rolling bearing again；Using Multiscale Morphological to PF points Amount is filtered, and the instantaneous amplitude of each PF components is then calculated with Teager energy operators, identifies the fault characteristic frequency of bearing.
The above method achieves certain effect in Rolling Bearing Fault Character extraction, but does not inquire into LMD methods in morning The applicability of phase failure.Since rolling bearing initial failure signal characteristic is fainter, and it is mingled with substantial amounts of noise, it is single to make It is difficult extraction fault signature with LMD methods.Related kurtosis deconvolution (Maximum Correlated Kurtosis Deconvolution, abbreviation MCKD) it is a kind of new Signal Pre-Processing Method by propositions such as McDonald, can be in noise reduction While specified periodic impulse signal can be strengthened.
The content of the invention
The object of the present invention is to provide a kind of Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD, for solving Certainly or mitigate the above problem.
To reach above-mentioned purpose, the technical solution adopted by the present invention is a kind of rolling bearing early stage based on MCKD and LMD Method for diagnosing faults, it includes
Step 1：Rolling bearing device is measured using acceleration transducer, obtains vibration acceleration signal；
Step 2：MCKD noise reductions are carried out to vibration acceleration signal, obtain the signal after noise reduction；
Step 3：LMD decomposition is carried out to the signal after MCKD noise reductions, obtains some PF components；
Step 4：Using related coefficient of the obtained PF components with decomposing front signal as criterion, unnecessary low frequency is rejected PF components, choose effective PF collection and carry out spectrum analysis, and Binding experiment extracts fault signature.
Further, in the step 2, the process steps that MCKD noise reductions are carried out to acceleration signal are as follows：
1) periodic signal ynRelated kurtosis is defined as：
In formula,For filter vector, T is signal ynCycle, L be wave filter length, M is carry digit；
2) solution procedure is equal to solution equation, orderThe result tried to achieve is with the shape of matrix Formula is expressed as
In formula：
3) the specific implementation process of MCKD algorithms is：
3.1) selection cycle T, filter length L and carry digit M；
3.2) signal is calculated's
3.3) filtered output signal is calculated
3.4) byCalculateWithDetermine the coefficient of wave filter
If 3.5) the Δ CK of signal before and after filteringm(T) be less than setting value when, stop recurrence, otherwise return the 3.3rd step after Continuous circulation.
Further, carry digit M value ranges are 1~7.
Further, in the step 3, LMD decomposition is carried out to the signal after MCKD noise reductions, obtains some PF components Process steps it is as follows：
1) assume that the signal after MCKD noise reductions is x (i), find out all Local Extremum n of signal x (i)i, obtain all phases The absolute value that adjacent Local Extremum average value and all adjacent Local Extremums are subtracted each other, and difference divided by 2, obtain miAnd ai：
2) and then by all adjacent miConnected with straight line, then be smoothed with moving average method, obtain part Mean function m11(t), envelope estimation function a is obtained with same method11(t)；
3) local mean value function m is subtracted from signal x (t)11(t), obtain：
h11(t)=x (t)-m11(t)
4) h is used again11(t) divided by envelope estimation function a11(t) with to h11(t) it is demodulated, obtains：
s11(t)=h11(t)/a11(t)
If s11(t) be not pure FM signal, i.e. its envelope estimation function a12(t) it is unsatisfactory for a12(t)=1, then by s11 (t) (1) is returned to as original signal to continue cycling through；In actual use, set a small variation Δ, when meet 1- Δs≤ a1nDuring≤1+ Δs, stop circulation；
5) all envelope estimation functions are multiplied to obtain envelope signal：
6) by envelope signal a1(t) and pure FM signal s1n(t) it is multiplied and obtains first PF component of original signal：
PF1(t)=a1(t)s1n
7) by PF1(t) separated from Setting signal x (t), obtain a new signal u1(t), by u1(t) as original Data duplication above step, is circulated k times, until uk(t) untill for a monotonic function.
8) so given original signal x (t) is broken down into k PF component and ukThe sum of (t), i.e.,
Further, in the step 4, using obtained PF components with decomposing the related coefficient of front signal as judgement Standard, is defined as x (n), single PF components are defined as y (n), and two signal x (n) are related to y's (n) by LMD decomposition front signals Coefficient ρxyIt is defined as follows:
In formula, correlation coefficient value ρxyValue between 0 and 1, ρxyCorrelation of the value between 0, two signals is just It is smaller, conversely, the correlation between two signals is bigger.Can be by ρxyValue is considered as pseudo- component than relatively low component.
The present invention based on the Fault Diagnosis of Rolling Element Bearings method of MCKD and LMD by using MCKD methods to roll Bearing acceleration signal is pre-processed, and while noise reduction, strengthens the impact component of signal, and by LMD methods to rolling Dynamic bearing acceleration signal is decomposed, any one complicated non-stationary signal adaptively is resolved into multiple instantaneous frequencys The sum of PF components with physical significance, each of which PF components are all by an envelope signal and a pure FM signal phase Multiplied to arrive, its instantaneous frequency has physical significance, while it is also a process progressively decomposed from high frequency to low frequency that LMD, which is decomposed, It is more accurate for fault diagnosis.
Brief description of the drawings
Attached drawing herein is merged in specification and forms the part of this specification, shows the implementation for meeting the present invention Example, and for explaining the principle of the present invention together with specification.
Fig. 1 is the diagnostic method flow chart of the present invention.
Fig. 2 is original bearing acceleration signal.
Fig. 3 is original bearing acceleration signal frequency spectrum.
Fig. 4 decomposes for original acceleration signal LMD.
Fig. 5 is PF1, PF2 component result of spectrum analysis.
Fig. 6 is MCKD de-noising signal time domain waveforms.
Fig. 7 decomposes for MCKD de-noising signals LMD.
Fig. 8 is that the signal LMD after MCKD noise reductions is decomposed.
Embodiment
To make the purpose, technical scheme and advantage that the present invention is implemented clearer, below in conjunction with the embodiment of the present invention Attached drawing, the technical solution in the embodiment of the present invention is further described in more detail.
The Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD of the present invention includes the following steps, such as Fig. 1 institutes Show：
Step 1：Rolling bearing device is measured using acceleration transducer, obtains vibration acceleration signal；
Step 2：To vibration acceleration signal carry out MCKD noise reductions, including the step of it is as follows：
1) signal ynRelated kurtosis is defined as：
In formula,For filter vector, T is signal ynCycle, L is the length of wave filter, and M is carry digit, and M values are generally 1~7 can be taken.
2) solution procedure is equal to solution equation, orderThe result tried to achieve is with the shape of matrix Formula is expressed as
In formula：
3) the specific implementation process of MCKD algorithms is：
3.1) selection cycle T, filter length L and carry digit M；
3.2) signal is calculated's
3.3) filtered output signal is calculated
3.4) byCalculateWithDetermine the coefficient of wave filter(5) if filtering before and after signal Δ CKm(T) it is less than During the value of setting, stop recurrence, otherwise return to (3) step and continue cycling through.
Step 3：LMD decomposition is carried out to the signal after MCKD noise reductions, some PF components is obtained, comprises the following steps that：
1) assume that the signal after MCKD noise reductions is x (i), find out all Local Extremum n of signal x (i)i, obtain all phases The absolute value that adjacent Local Extremum average value and all adjacent Local Extremums are subtracted each other, and difference divided by 2, obtain miAnd ai。
2) and then by all adjacent miConnected with straight line, then be smoothed with moving average method, obtain part Mean function m11(t).Envelope estimation function a is obtained with same method11(t)。
3) local mean value function m is subtracted from signal x (t)11(t), obtain：
h11(t)=x (t)-m11(t)
4) h is used again11(t) divided by envelope estimation function a11(t) with to h11(t) it is demodulated, obtains：
s11(t)=h11(t)/a11(t)
If s11(t) be not pure FM signal, i.e. its envelope estimation function a12(t) it is unsatisfactory for a12(t)=1, then by s11 (t) (1) is returned to as original signal to continue cycling through.In actual use, set a small variation Δ, when meet 1- Δs≤ a1nDuring≤1+ Δs, stop circulation.
5) all envelope estimation functions are multiplied to obtain envelope signal：
6) by envelope signal a1(t) and pure FM signal s1n(t) it is multiplied and obtains first PF component of original signal：
PF1(t)=a1(t)s1n
7) by PF1(t) separated from Setting signal x (t), obtain a new signal u1(t), by u1(t) as original Data duplication above step, is circulated k times, until uk(t) untill for a monotonic function.
8) so given original signal x (t) is broken down into k PF component and ukThe sum of (t), i.e.,
Step 4：Using related coefficient of the obtained PF components with decomposing front signal as criterion, unnecessary low frequency is rejected PF components, choose effective PF collection and carry out spectrum analysis, and Binding experiment extracts fault signature.
With one group of concrete numerical value, the invention will be further described：Set bearing fault simulation test bed rotor speed as 1797rpm/min, sample frequency 12KHz, outer ring failure-frequency fo≈ 107.82Hz, shaft fundamental frequency are fr=29.95Hz.Adopt Collect one group of bearing outer ring fault vibration acceleration signal and see Fig. 2, while the spectrogram for providing the signal is shown in Fig. 3, can be with from figure Find out and contain 4 times of shaft fundamental frequencies and 2 times of outer ring failure-frequencies in signal, but amplitude is smaller, it is difficult to find out containing outer ring failure frequency Rate 107.82Hz, with there is interference spectral line excessively unrelated with characteristic frequency in time-frequency spectrum, fault signature is fainter.
Fig. 4 be LMD decomposition directly is carried out to primary fault acceleration signal as a result, table 1 give each PF components with The related coefficient of primary fault acceleration signal, takes the larger PF1 and PF2 components of related coefficient to carry out result of spectrum analysis as schemed Shown in 4, there are rotating speed fundamental frequency and fault characteristic frequency on PF1 and PF2 component spectrograms.In PF2 component maps, in failure-frequency 2 frequencys multiplication at also have a faint peak value, but same overall see fault characteristic frequency peak value and unobvious.
The related coefficient of each PF components of table 1 and fault-signal
PF1 PF2 PF3 PF4
0.817 0.5810 0.2082 0.0982
Failure acceleration is analyzed using method proposed by the present invention, primary fault is believed using MCKD methods first Number carrying out noise reduction obtains that the results are shown in Figure 5, as can be seen from Figure 5 substantially occurs many impact components in signal.It is right Signal after noise reduction carries out LMD and decomposes to obtain PF components as shown in fig. 6, compared with Fig. 3, and PF components impact component showed increased, And there is certain regularity.
The related coefficient of each PF components of table 2 and signal after MCKD noise reductions
PF1 PF2 PF3 PF4
0.8465 0.5863 0.3603 0.2383
The related coefficient provided according to table 2, equally takes larger PF1, PF2 and PF3 component of related coefficient to carry out frequency spectrum point Analysis, obtained spectrogram on PF1 and PF2 components as shown in fig. 7, it can be seen from the figure that can clearly find to turn fundamental frequency There is apparent peak value at 2 frequency multiplication of rate, failure-frequency and failure-frequency.There is also failure-frequency and its 2 times on PF3 spectrograms Frequency component.Compared with Fig. 4, width at failure-frequency point on PF1 components in Amplitude Ration Fig. 4 at the PF1 components failure-frequency point in Fig. 7 Value adds about 1.6 times, and also has obvious peak value at 2 frequency multiplication of failure-frequency.Moreover, in Fig. 7 PF2 components failure All than the amplitude of this Frequency point at two adds about 1 times on PF2 components in Fig. 4, failure is special for amplitude at Frequency point and its 2 frequency multiplication Sign frequency peak is more protruded, and failure is more easy to find.Thus can be easier to determine that housing washer has damage, the result drawn It is consistent with truth.
To sum up, it is of the invention based on the Fault Diagnosis of Rolling Element Bearings method of MCKD and LMD by using MCKD methods Rolling bearing acceleration signal is pre-processed, while noise reduction, strengthens the impact component of signal, and pass through LMD side Method decomposes rolling bearing acceleration signal, any one complicated non-stationary signal adaptively is resolved into multiple winks When frequency there is the sum of PF components of physical significance, each of which PF components are all by an envelope signal and a pure frequency modulation Signal multiplication obtains, its instantaneous frequency has physical significance, while LMD decomposition is also one and is progressively decomposed from high frequency to low frequency Process, it is more accurate for fault diagnosis.
The above, is only the optimal embodiment of the present invention, but protection scope of the present invention is not limited thereto, Any one skilled in the art the invention discloses technical scope in, the change or replacement that can readily occur in, It should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with the protection model of the claim Subject to enclosing.

Claims (5)

  1. A kind of 1. Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD, it is characterised in that including
    Step 1：Rolling bearing device is measured using acceleration transducer, obtains vibration acceleration signal；
    Step 2：MCKD noise reductions are carried out to vibration acceleration signal, obtain the signal after noise reduction；
    Step 3：LMD decomposition is carried out to the signal after MCKD noise reductions, obtains some PF components；
    Step 4：Using related coefficient of the obtained PF components with decomposing front signal as criterion, unnecessary low frequency PF points are rejected Amount, choose effective PF collection and carry out spectrum analysis, and Binding experiment extracts fault signature.
  2. 2. the Fault Diagnosis of Rolling Element Bearings method according to claim 1 based on MED and cepstrum, its feature exist In in the step 2, the process steps that MCKD noise reductions are carried out to acceleration signal are as follows：
    1) periodic signal ynRelated kurtosis is defined as：
    <mrow> <msub> <mi>CK</mi> <mi>M</mi> </msub> <mrow> <mo>(</mo> <mi>T</mi> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>max</mi> <mover> <mi>f</mi> <mo>&amp;RightArrow;</mo> </mover> </munder> <mfrac> <mrow> <munderover> <mi>&amp;Sigma;</mi> <mrow> <mi>n</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <msup> <mrow> <mo>(</mo> <mrow> <msubsup> <mi>&amp;Pi;</mi> <mrow> <mi>m</mi> <mo>=</mo> <mn>0</mn> </mrow> <mi>M</mi> </msubsup> <msub> <mi>y</mi> <mrow> <mi>n</mi> <mo>-</mo> <mi>m</mi> <mi>T</mi> </mrow> </msub> </mrow> <mo>)</mo> </mrow> <mn>2</mn> </msup> </mrow> <msup> <mrow> <mo>(</mo> <mrow> <munderover> <mi>&amp;Sigma;</mi> <mrow> <mi>n</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <msubsup> <mi>y</mi> <mi>n</mi> <mn>2</mn> </msubsup> </mrow> <mo>)</mo> </mrow> <mrow> <mi>M</mi> <mo>+</mo> <mn>1</mn> </mrow> </msup> </mfrac> <mo>,</mo> <mover> <mi>f</mi> <mo>&amp;RightArrow;</mo> </mover> <mo>=</mo> <msup> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <msub> <mi>f</mi> <mn>1</mn> </msub> </mtd> <mtd> <msub> <mi>f</mi> <mn>2</mn> </msub> </mtd> <mtd> <mn>...</mn> </mtd> <mtd> <msub> <mi>f</mi> <mi>L</mi> </msub> </mtd> </mtr> </mtable> </mfenced> <mi>T</mi> </msup> </mrow>
    In formula,For filter vector, T is signal ynCycle, L be wave filter length, M is carry digit；
    2) solution procedure is equal to solution equation, orderK=1,2 ..., L, the result tried to achieve is with a matrix type It is expressed as
    In formula：R=[0 T 2T ... mT]
    <mrow> <mover> <msub> <mi>&amp;alpha;</mi> <mi>m</mi> </msub> <mo>&amp;RightArrow;</mo> </mover> <mo>=</mo> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <mrow> <msubsup> <mi>y</mi> <mrow> <mn>1</mn> <mo>-</mo> <mi>m</mi> <mi>T</mi> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <mrow> <msubsup> <mi>y</mi> <mn>1</mn> <mn>2</mn> </msubsup> <msubsup> <mi>y</mi> <mrow> <mn>1</mn> <mo>-</mo> <mi>T</mi> </mrow> <mn>2</mn> </msubsup> <mn>...</mn> <msubsup> <mi>y</mi> <mrow> <mn>1</mn> <mo>-</mo> <mi>M</mi> <mi>T</mi> </mrow> <mn>2</mn> </msubsup> </mrow> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msubsup> <mi>y</mi> <mrow> <mn>2</mn> <mo>-</mo> <mi>m</mi> <mi>T</mi> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <mrow> <msubsup> <mi>y</mi> <mn>2</mn> <mn>2</mn> </msubsup> <msubsup> <mi>y</mi> <mrow> <mn>2</mn> <mo>-</mo> <mi>T</mi> </mrow> <mn>2</mn> </msubsup> <mn>...</mn> <msubsup> <mi>y</mi> <mrow> <mn>2</mn> <mo>-</mo> <mi>M</mi> <mi>T</mi> </mrow> <mn>2</mn> </msubsup> </mrow> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> <mtr> <mtd> <mtable> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> </mtable> </mtd> </mtr> <mtr> <mtd> <mrow> <msubsup> <mi>y</mi> <mrow> <mi>N</mi> <mo>-</mo> <mi>m</mi> <mi>T</mi> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <mrow> <msubsup> <mi>y</mi> <mi>N</mi> <mn>2</mn> </msubsup> <msubsup> <mi>y</mi> <mrow> <mi>N</mi> <mo>-</mo> <mi>T</mi> </mrow> <mn>2</mn> </msubsup> <mn>...</mn> <msubsup> <mi>y</mi> <mrow> <mi>N</mi> <mo>-</mo> <mi>M</mi> <mi>T</mi> </mrow> <mn>2</mn> </msubsup> </mrow> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>;</mo> <mover> <mi>&amp;beta;</mi> <mo>&amp;RightArrow;</mo> </mover> <mo>=</mo> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <mrow> <msub> <mi>y</mi> <mn>1</mn> </msub> <msub> <mi>y</mi> <mrow> <mn>1</mn> <mo>-</mo> <mi>T</mi> </mrow> </msub> <mn>...</mn> <msub> <mi>y</mi> <mrow> <mn>1</mn> <mo>-</mo> <mi>M</mi> <mi>T</mi> </mrow> </msub> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>y</mi> <mn>2</mn> </msub> <msub> <mi>y</mi> <mrow> <mn>2</mn> <mo>-</mo> <mi>T</mi> </mrow> </msub> <mn>...</mn> <msub> <mi>y</mi> <mrow> <mn>2</mn> <mo>-</mo> <mi>M</mi> <mi>T</mi> </mrow> </msub> </mrow> </mtd> </mtr> <mtr> <mtd> <mtable> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> </mtable> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>y</mi> <mi>N</mi> </msub> <msub> <mi>y</mi> <mrow> <mi>N</mi> <mo>-</mo> <mi>T</mi> </mrow> </msub> <mn>...</mn> <msub> <mi>y</mi> <mrow> <mi>N</mi> <mo>-</mo> <mi>M</mi> <mi>T</mi> </mrow> </msub> </mrow> </mtd> </mtr> </mtable> </mfenced> </mrow>
    3) the specific implementation process of MCKD algorithms is：
    3.1) selection cycle T, filter length L and carry digit M；
    3.2) signal is calculated's
    3.3) filtered output signal is calculated
    3.4) byCalculateWithDetermine the coefficient of wave filter
    If 3.5) the Δ CK of signal before and after filteringm(T) when being less than the value of setting, stop recurrence, otherwise return to the 3.3rd step and continue to follow Ring.
  3. 3. the Fault Diagnosis of Rolling Element Bearings method according to claim 2 based on MED and cepstrum, its feature exist In carry digit M value ranges are 1~7.
  4. 4. the Fault Diagnosis of Rolling Element Bearings method according to claim 3 based on MED and cepstrum, its feature exist In in the step 3, to the signal progress LMD decomposition after MCKD noise reductions, the process steps for obtaining some PF components are as follows：
    1) assume that the signal after MCKD noise reductions is x (i), find out all Local Extremum n of signal x (i)i, obtain all adjacent parts The absolute value that extreme point average value and all adjacent Local Extremums are subtracted each other, and difference divided by 2, obtain miAnd ai：
    2) and then by all adjacent miConnected with straight line, then be smoothed with moving average method, obtain local mean value letter Number m11(t), envelope estimation function a is obtained with same method11(t)；
    3) local mean value function m is subtracted from signal x (t)11(t), obtain：
    h11(t)=x (t)-m11(t)
    4) h is used again11(t) divided by envelope estimation function a11(t) with to h11(t) it is demodulated, obtains：
    s11(t)=h11(t)/a11(t)
    If s11(t) be not pure FM signal, i.e. its envelope estimation function a12(t) it is unsatisfactory for a12(t)=1, then by s11(t) make (1) is returned to for original signal to continue cycling through；In actual use, a small variation Δ is set, when meeting 1- Δs≤a1n≤1 During+Δ, stop circulation；
    5) all envelope estimation functions are multiplied to obtain envelope signal：
    <mrow> <msub> <mi>a</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mi>a</mi> <mn>11</mn> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <msub> <mi>a</mi> <mn>12</mn> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>...</mo> <msub> <mi>a</mi> <mrow> <mn>1</mn> <mi>n</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <munderover> <mo>&amp;Pi;</mo> <mrow> <mi>q</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>a</mi> <mrow> <mn>1</mn> <mi>q</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow>
    6) by envelope signal a1(t) and pure FM signal s1n(t) it is multiplied and obtains first PF component of original signal：
    PF1(t)=a1(t)s1n
    7) by PF1(t) separated from Setting signal x (t), obtain a new signal u1(t), by u1(t) it is used as initial data Above step is repeated, is circulated k times, until uk(t) untill for a monotonic function.
    8) so given original signal x (t) is broken down into k PF component and ukThe sum of (t), i.e.,
    <mrow> <mi>x</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>p</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>k</mi> </munderover> <msub> <mi>PF</mi> <mi>p</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>+</mo> <msub> <mi>u</mi> <mi>k</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>.</mo> </mrow>
  5. 5. the Fault Diagnosis of Rolling Element Bearings method according to claim 4 based on MED and cepstrum, its feature exist In, in the step 4, using obtained PF components with decompose front signal related coefficient as criterion, LMD is decomposed Front signal is defined as x (n), and single PF components are defined as the correlation coefficient ρ of y (n), two signal x (n) and y (n)xyDefinition such as Under:
    <mrow> <msub> <mi>&amp;rho;</mi> <mrow> <mi>x</mi> <mi>y</mi> </mrow> </msub> <mo>=</mo> <mo>|</mo> <mfrac> <mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>n</mi> <mo>=</mo> <mn>0</mn> </mrow> <mi>&amp;infin;</mi> </munderover> <mi>x</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <mi>y</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </mrow> <msqrt> <mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>n</mi> <mo>=</mo> <mn>0</mn> </mrow> <mi>&amp;infin;</mi> </munderover> <msup> <mi>x</mi> <mn>2</mn> </msup> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>n</mi> <mo>=</mo> <mn>0</mn> </mrow> <mi>&amp;infin;</mi> </munderover> <msup> <mi>y</mi> <mn>2</mn> </msup> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </mrow> </msqrt> </mfrac> <mo>|</mo> </mrow>
    In formula, correlation coefficient value ρxyValue between 0 and 1, ρxyIt is worth the correlation between 0, two signals with regard to smaller, Conversely, the correlation between two signals is bigger.Can be by ρxyValue is considered as pseudo- component than relatively low component.
CN201711256531.7A 2017-12-03 2017-12-03 A kind of Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD Pending CN107917806A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711256531.7A CN107917806A (en) 2017-12-03 2017-12-03 A kind of Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711256531.7A CN107917806A (en) 2017-12-03 2017-12-03 A kind of Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD

Publications (1)

Publication Number Publication Date
CN107917806A true CN107917806A (en) 2018-04-17

Family

ID=61898314

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711256531.7A Pending CN107917806A (en) 2017-12-03 2017-12-03 A kind of Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD

Country Status (1)

Country Link
CN (1) CN107917806A (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108664936A (en) * 2018-05-14 2018-10-16 浙江师范大学 A kind of diagnostic method and system based on mechanical disorder
CN108716463A (en) * 2018-04-19 2018-10-30 合肥通用机械研究院有限公司 A kind of method for diagnosing faults of reciprocating compressor ring air flap
CN108844740A (en) * 2018-06-27 2018-11-20 中国直升机设计研究所 A kind of Fault Diagnosis of Rolling Element Bearings method based on MED and cepstrum
CN109029977A (en) * 2018-07-12 2018-12-18 福州大学 A kind of epicyclic gearbox Incipient Fault Diagnosis method based on VMD-AMCKD
CN109063672A (en) * 2018-08-21 2018-12-21 北京工业大学 A kind of early stage bearing outer ring method for diagnosing faults based on adaptive M CKD
CN109682601A (en) * 2019-03-04 2019-04-26 北京天泽智云科技有限公司 The initial failure recognition methods of rolling bearing under a kind of variable speed operating condition
CN110044608A (en) * 2019-04-19 2019-07-23 石家庄铁道大学 Vibration signal compression sensing method based on MCKD
CN111504645A (en) * 2020-04-30 2020-08-07 西安交通大学 Rolling bearing fault diagnosis method based on frequency domain multipoint kurtosis
CN111896260A (en) * 2020-08-01 2020-11-06 华东交通大学 NGAs synchronous optimization wavelet filter and MCKD bearing fault diagnosis method

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104236905A (en) * 2014-08-26 2014-12-24 中国直升机设计研究所 Bearing fault diagnosis method
CN105628381A (en) * 2016-01-13 2016-06-01 东北石油大学 Reciprocating compressor bearing fault diagnosis method based on improved local mean value decomposition
CN106908241A (en) * 2017-02-23 2017-06-30 北京工业大学 A kind of bearing fault method of discrimination being combined with Wavelet Denoising Method based on LMD

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104236905A (en) * 2014-08-26 2014-12-24 中国直升机设计研究所 Bearing fault diagnosis method
CN105628381A (en) * 2016-01-13 2016-06-01 东北石油大学 Reciprocating compressor bearing fault diagnosis method based on improved local mean value decomposition
CN106908241A (en) * 2017-02-23 2017-06-30 北京工业大学 A kind of bearing fault method of discrimination being combined with Wavelet Denoising Method based on LMD

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
任学平 等: "基于 MCKD-EEMD 的滚动轴承微故障特征提取", 《机械设计与制造》 *
孙伟 等: "小波包降噪与 LMD 相结合的滚动轴承故障诊断方法", 《振动与冲击》 *

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108716463A (en) * 2018-04-19 2018-10-30 合肥通用机械研究院有限公司 A kind of method for diagnosing faults of reciprocating compressor ring air flap
CN108664936A (en) * 2018-05-14 2018-10-16 浙江师范大学 A kind of diagnostic method and system based on mechanical disorder
CN108664936B (en) * 2018-05-14 2020-09-01 浙江师范大学 Diagnosis method and system based on machine fault
CN108844740A (en) * 2018-06-27 2018-11-20 中国直升机设计研究所 A kind of Fault Diagnosis of Rolling Element Bearings method based on MED and cepstrum
CN109029977A (en) * 2018-07-12 2018-12-18 福州大学 A kind of epicyclic gearbox Incipient Fault Diagnosis method based on VMD-AMCKD
CN109063672A (en) * 2018-08-21 2018-12-21 北京工业大学 A kind of early stage bearing outer ring method for diagnosing faults based on adaptive M CKD
CN109682601A (en) * 2019-03-04 2019-04-26 北京天泽智云科技有限公司 The initial failure recognition methods of rolling bearing under a kind of variable speed operating condition
CN109682601B (en) * 2019-03-04 2020-11-17 北京天泽智云科技有限公司 Early fault identification method for rolling bearing under variable rotating speed working condition
CN110044608A (en) * 2019-04-19 2019-07-23 石家庄铁道大学 Vibration signal compression sensing method based on MCKD
CN111504645A (en) * 2020-04-30 2020-08-07 西安交通大学 Rolling bearing fault diagnosis method based on frequency domain multipoint kurtosis
CN111896260A (en) * 2020-08-01 2020-11-06 华东交通大学 NGAs synchronous optimization wavelet filter and MCKD bearing fault diagnosis method
CN111896260B (en) * 2020-08-01 2022-05-13 华东交通大学 NGAs synchronous optimization wavelet filter and MCKD bearing fault diagnosis method

Similar Documents

Publication Publication Date Title
CN107917806A (en) A kind of Fault Diagnosis of Rolling Element Bearings method based on MCKD and LMD
Zheng et al. Incipient fault detection of rolling bearing using maximum autocorrelation impulse harmonic to noise deconvolution and parameter optimized fast EEMD
Li et al. Rolling bearing fault diagnosis based on improved adaptive parameterless empirical wavelet transform and sparse denoising
CN104236905A (en) Bearing fault diagnosis method
CN108168891B (en) Method and equipment for extracting weak fault signal characteristics of rolling bearing
Li et al. An enhanced rolling bearing fault detection method combining sparse code shrinkage denoising with fast spectral correlation
CN104156585B (en) A kind of doubledictionary match tracing bearing fault degree evaluation method based on many index feature
CN106908241A (en) A kind of bearing fault method of discrimination being combined with Wavelet Denoising Method based on LMD
CN109883706B (en) Method for extracting local damage weak fault features of rolling bearing
CN108801630B (en) Gear fault diagnosis method for single-channel blind source separation
CN111238813B (en) Method for extracting fault features of rolling bearing under strong interference
CN112101245A (en) Short-time Fourier transform mechanical impact feature extraction method based on frequency domain window function
CN105628381A (en) Reciprocating compressor bearing fault diagnosis method based on improved local mean value decomposition
Liu et al. Bearing fault diagnosis method with unknown variable speed based on multi-curve extraction and selection
CN111769810B (en) Fluid mechanical modulation frequency extraction method based on energy kurtosis spectrum
CN104215459A (en) Bearing fault diagnosis method
CN106409351A (en) Method and device for processing vibration signal of loose part of nuclear power plant
CN110940524B (en) Bearing fault diagnosis method based on sparse theory
CN109374298B (en) Bearing fault diagnosis method based on cross-correlation singular value
CN114992033B (en) NLM-CEEMDAN-based hydroelectric generating set signal denoising method
CN111504640B (en) Weighted sliding window second-order synchronous compression S transformation bearing fault diagnosis method
Zhang et al. Fault diagnosis for gearbox based on EMD-MOMEDA
CN106908232A (en) A kind of vibration signal fault recognition method of rotating machinery
CN108398260B (en) Method for quickly evaluating instantaneous angular speed of gearbox based on mixed probability method
CN106706122A (en) Correlation coefficient and EMD (Empirical Mode Decomposition) filtering characteristic-based rub-impact acoustic emission signal noise reduction method

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20180417