CN102778357B - Mechanical failure feature extracting method based on optimal parameter ensemble empirical mode decomposition (EEMD) - Google Patents

Abstract

The invention provides a mechanical failure feature extracting method based on optimal parameter ensemble empirical mode decomposition (EEMD), relating to the technical field of mechanical equipment state monitoring and failure diagnosis. According to the mechanical failure feature extracting method, based on the rule that signal extreme point distribution uniformity is changed by a white noise amplitude coefficient, a white noise amplitude coefficient enabling the best signal extreme point uniformity is used as a k value of EEMD, and an M value (total average number) is obtained through calculation by setting an expected decomposition error, thus the EEMD parameter is selected, the EEMD is carried out on a failure signal, and a failure feature is extracted. The feasibility and the effectiveness of the mechanical failure feature extracting method are verified by the successful extraction of the failure feature of a rolling bearing. The mechanical failure feature extracting method is suitable for the fields of mechanical equipment state monitoring, failure diagnosis and the like.

Description

Mechanical fault feature extracting method based on optimized parameter set empirical mode decomposition

Technical field

The present invention relates to mechanical equipment state monitoring and fault diagnosis technology field, be specifically related to a kind of mechanical equipment fault feature extracting method.

Background technology

Set empirical mode decomposition (Ensemble empirical mode decomposition, EEMD) noise assistant analysis is applied to empirical mode decomposition (Empirical Mode decompos it ion, EMD) in, to promote anti-mixed decomposition, effectively suppress pattern intrinsic in EMD and obscure problem.With respect to EMD, through EEMD, decompose the physical connotation that the intrinsic mode function (Intrinsic mode function, IMF) obtaining more can disclose original signal, make the physical essence of each IMF more clear.Lei is applied to EEMD in the fault diagnosis of rotary machine rotor system, Accurate Diagnosis generator amature touch the fault of rubbing, and proved EEMD superiority in rotating machinery fault feature extraction; Zvokelj combines EEMD and principal component analysis, has improved the reliability of rolling bearing fault diagnosis; The loosening fault signature of wind power generating set bearing seat has effectively been extracted in An application EEMD and Hilbert transform; Zhou application EEMD processes the signal that gear case embedded type sensor obtains, and has successfully realized gear case state on_line monitoring and fault diagnosis; Lei Yaguo has proposed the mechanical failure diagnostic method based on improving Hilbert-Huang conversion, first utilize EEMD to obtain the IMF that non-mode is obscured, by susceptibility assessment algorithm, select the IMF to fault signature sensitivity again, make the Hilbert-Huang spectrum obtaining can diagnose more accurately mechanical fault.

More than research has shown that EEMD has important value and clear superiority in mechanical fault feature extraction, yet, when application EEMD, two important parameters must be set, add amplitude coefficient k and the population mean number of times M of white noise.If it is improper that these two parameters arrange, can make resolution error increase, cause decomposition result meaningless.When k is too small, may be not enough to cause the variation of signal Local Extremum, make to add noise to lose meaning to change local time's span of signal, when k is excessive, can make resolution error increase, even can fall into oblivion original signal feature and make decomposition lose meaning.Theoretically, M value more resolution error is less, until ignore, but the increase of M will be lost counting yield, make consuming time being multiplied.On The Choice for k and M, Huang suggestion k is multiplied by a mark by the standard deviation of original signal and defines, and generally get original signal standard deviation 0.2 times, signal medium-high frequency composition k of many times suitably reduces, otherwise suitably increase k, M can determine by resolution error is set; Recently definite k of the amplitude standard deviation of radio-frequency component in old summary application signal and the amplitude standard deviation of low-frequency component, by the expectation resolution error of setting, obtain M again, the method has certain using value, but radio-frequency component and the low-frequency component of the signal that is not easily distinguishable in computation process.

Summary of the invention

In view of this, the present invention is directed to two important parameter k in EEMD and the On The Choice of M, analyze white noise amplitude coefficient and changed the rule that original signal extreme point distributes, and the impact on EEMD Decomposition Accuracy, efficiency and resolution error, by calculating, add the distributing homogeneity index of extreme point after white noise to determine k, the expectation resolution error that EEMD is set calculates M.Thereby, the EEMD method of optimized parameter has been proposed, so that EEMD is for mechanical movement status monitoring or carry out fault diagnosis.

The mechanical fault feature extracting method based on optimized parameter set empirical mode decomposition providing of the present invention, comprises the steps:

1) input signal x (t), calculating adds the product std_max of the maximum value sequence consecutive point amplitude difference of signal and the standard deviation of spacing after the white noise of different amplitude coefficients, and the product std_min of the standard deviation of minimal value sequence consecutive point amplitude difference and spacing, using this index as the extremely large and small value point distributing homogeneity of evaluation signal, the large and small value of the less pickup electrode of std_max and std_min distributes more even, acquisition makes std_max obtain the white noise amplitude coefficient k_max of minimum value, obtains and makes std_min obtain the white noise amplitude coefficient k_min of minimum value;

2) ask the mean value of k_max and k_min as the parameter k of EEMD, then according to the expectation resolution error e calculated population average time M arranging, complete EEMD parameter optimization;

3) fault-signal is carried out to EEMD decomposition, obtain a series of intrinsic mode function IMF components, complete fault signature and extract.

Further, described step 1) specifically comprises the steps:

11) obtain the standard deviation of original signal x (t), add respectively the white noise n' of N different amplitude coefficients _i(t), white noise amplitude coefficient is shown below:

$k\left(i\right)=\frac{i}{100}\mathrm{\σ}(i=1~N),$

Wherein, k (i) represents white noise amplitude coefficient, and i represents the i time and adds white noise;

12) calculate the upper extreme point sequence extr_max (x1 at every turn add signal after white noise, y1), calculate the lower extreme point sequence extr_min (x2, y2) at every turn add signal after white noise, wherein, (x1, y1), (x2, y2) are horizontal stroke, the ordinate of corresponding extreme point;

13) calculate amplitude difference standard deviation and the poor long-pending std_max of separation criteria of extr_max (x1, y1) consecutive point, calculate amplitude difference standard deviation and the poor long-pending std_min of separation criteria of extr_min (x2, y2) consecutive point, as shown in the formula:

$\mathrm{std}\_\max =\sqrt{\frac{1}{N1-1}\underset{i=N1}{\overset{N1-1}{\mathrm{\Σ}}}\left(\right|{x1}_{i+1}-{x1}_i|-u_1)}\×\sqrt{\frac{1}{N1-1}\underset{i=N1}{\overset{N1-1}{\mathrm{\Σ}}}\left(\right|{y1}_{i+1}-{y1}_i|-u_1^{\′})}$

$\mathrm{std}\_\min =\sqrt{\frac{1}{N2-1}\underset{i=N2}{\overset{N2-1}{\mathrm{\Σ}}}\left(\right|{x2}_{i+1}-{x2}_i|-u_2)}\×\sqrt{\frac{1}{N2-1}\underset{i=N2}{\overset{N2-1}{\mathrm{\Σ}}}\left(\right|{y2}_{i+1}-{y2}_i\left|{-u}_2^{\′}\right)}$

In formula, N1 is counting of extr_max (x1, y1), and N2 is that (x 2, counting y2), u for extr_min ₁, u ₁' be the mean value of extr_max (x1, y1) consecutive point amplitude difference and the mean value of spacing, u ₂, u ' ₂for the mean value of extr_min (x2, y2) consecutive point amplitude difference and the mean value of spacing, x1 _i+1represent maximum point horizontal ordinate, x1 _irepresent maximum point horizontal ordinate, y1 _i+1represent maximum point ordinate, y1 _irepresent maximum point ordinate, x2 _i+1represent minimum point horizontal ordinate, x2 _irepresent minimum point horizontal ordinate, y2 _i+1represent minimum point ordinate, y2 _irepresent minimum point ordinate.

Further, described step 2) specifically comprise the steps:

14) calculate and make std_max and std_min reach the white noise amplitude coefficient k_max of minimum value and the mean value k of k_min, then calculate M according to following formula:

$M={\left(\frac{k}{e}\right)}^2,$

And whether the size that judges M be less than 20, if during M < 20, get M=20.

Further, described step 3) specifically comprises the steps:

15) in original vibration signal x (t), adding respectively in such a way M average is 0, the white Gaussian noise n that amplitude coefficient is k _i(t):

x _i?(t)＝x(t)＋n _i?(t)，

In formula, i=1～M, i represents the i time and adds white noise, x _i(t) for adding the signal after white Gaussian noise;

16) to x _i(t) gather respectively empirical mode decomposition, obtain K intrinsic mode function component and a remainder r _i(t):

$x_i\left(t\right)=\underset{j=1}{\overset{K}{\mathrm{\&Sigma;}}}{c}_{\mathrm{ij}}\left(t\right)+r_i\left(t\right)$

In formula, c _ij(t) represent to add for the i time after white Gaussian noise, decompose resulting j intrinsic mode function component, j=1～K;

17) by above-mentioned steps 16) corresponding intrinsic mode function component carries out population mean computing, obtains gathering the intrinsic mode function component c after empirical mode decomposition _jand remainder r (t) (t):

$c_j\left(t\right)=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{c}_{\mathrm{ij}}\left(t\right),$

$r\left(t\right)=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{r}_i\left(t\right),$

In formula, c _j(t) represent original signal to gather resulting j intrinsic mode function component after empirical mode decomposition;

18) obtain intrinsic mode function component and a remainder of K the ascending ordered arrangement of characteristic time scale:

$x\left(t\right)=\underset{j=1}{\overset{K}{\mathrm{\&Sigma;}}}{c}_j\left(t\right)+r\left(t\right),$

Wherein, x (t) represents original signal, c _j(t) represent j IMF component, r (t) represents remainder.

Further, the fault signature in described step 3) is that the amplitude spectrum of the IMF component after decomposing by EEMD and Hilbert have composed fault signature and extract.

Further, described error e generally gets 0.01.

Further, described white noise n' _i(t) amplitude coefficient N generally gets 50.

The present invention changes signal extreme point distributing homogeneity rule and the rule to EEMD resolution error and Decomposition Accuracy from the white noise of different amplitude coefficients, searching makes the most uniform white noise amplitude coefficient of signal extreme point distribution as the k value of EEMD, at the expectation resolution error by arranging, carry out calculated population average time M, thereby complete the optimization of EEMD parameter.By the parameter after optimizing, fault-signal is carried out to EEMD decomposition, obtain a series of IMF component, observe the IMF component that characterizes fault signature, or ask for corresponding amplitude spectrum and Hilbert spectrum extraction fault signature.The present invention is for application EEMD is while carrying out mechanical fault feature extraction, parameter a kind of new method that provides is provided, the method is from the feature of signal own, the adaptive EEMD parameter that is applicable to signal to be analyzed of choosing.

Other advantages of the present invention, target, to set forth in the following description to a certain extent with feature, and based on will be apparent to those skilled in the art to investigating below, or can be instructed from the practice of the present invention to a certain extent.Target of the present invention and other advantages can be passed through instructions, claims, and in accompanying drawing, specifically noted structure realizes and obtains.

Accompanying drawing explanation

Fig. 1 is algorithm flow chart of the present invention;

Fig. 2 is original signal;

Fig. 3 is the Fourier spectrum of original signal;

Fig. 4 is first rank IMF component-IMF1 after EEMD decomposes;

Fig. 5 is the amplitude spectrum of IMF1.

Embodiment

Below with reference to accompanying drawing, the preferred embodiments of the present invention are described in detail; Should be appreciated that preferred embodiment is only for the present invention is described, rather than in order to limit the scope of the invention.

Below will be described in detail the preferred embodiments of the present invention.

Referring to Fig. 1, Fig. 1 is algorithm flow chart of the present invention, and as shown in the figure, the mechanical fault feature extracting method based on optimized parameter set empirical mode decomposition provided by the invention, comprises the following steps:

1) obtain the standard deviation of original signal x (t), add respectively the white noise n' of N (N generally gets 50, and N can suitably increase if needed) different amplitude coefficients _i(t), white noise amplitude coefficient is shown below:

wherein, k (i) represents white noise amplitude coefficient, and i represents the i time and adds white noise;

2) calculate upper and lower extreme point sequence extr_max (x1, y1) and the extr_min (x2, y2) at every turn add signal after white noise, x1, y1(x2, y2) be horizontal stroke, the ordinate of corresponding extreme point;

3) calculate amplitude difference standard deviation and poor long-pending std_max and the std_min of separation criteria of extr_max (x1, y1) and extr_min (x2, y2) consecutive point, as shown in the formula:

$\mathrm{std}\_\max =\sqrt{\frac{1}{N1-1}\underset{i=N1}{\overset{N1-1}{\mathrm{\&Sigma;}}}\left(\right|{x1}_{i+1}-{x1}_i|-u_1)}\&times;\sqrt{\frac{1}{N1-1}\underset{i=N1}{\overset{N1-1}{\mathrm{\&Sigma;}}}\left(\right|{y1}_{i+1}-{y1}_i|-u_1^{\&prime;})},$

$\mathrm{std}\_\min =\sqrt{\frac{1}{N2-1}\underset{i=N2}{\overset{N2-1}{\mathrm{\&Sigma;}}}\left(\right|{x2}_{i+1}-{x2}_i|-u_2)}\&times;\sqrt{\frac{1}{N2-1}\underset{i=N2}{\overset{N2-1}{\mathrm{\&Sigma;}}}\left(\right|{y2}_{i+1}-{y2}_i|-u_2^{\&prime;})};$

In formula, N1, N2 are that (x 2, counting y2), u for extr_max (x1, y1) and extr_min ₁, u ₁', u ₂, u ' ₂(x 1, y1) and the mean value of extr_min (x2, y2) consecutive point amplitude difference and the mean value of spacing to be respectively extr_max; X1 _i+1represent maximum point horizontal ordinate, x1 _irepresent maximum point horizontal ordinate, y1 _i+1represent maximum point ordinate, y1 _irepresent maximum point ordinate, x2 _i+1represent minimum point horizontal ordinate, x2 _irepresent minimum point horizontal ordinate, y2 _i+1represent minimum point ordinate, y2 _irepresent minimum point ordinate.

4) calculate and make std_max and std_min reach the white noise amplitude coefficient k_max of minimum value and the mean value k of k_min, expectation resolution error e=0.01 is set, then calculates M according to following formula, during as M < 20, get M=20;

$M={\left(\frac{k}{e}\right)}^2;$

5) in original vibration signal x (t), adding respectively M average is 0, the white Gaussian noise n that amplitude coefficient is k _i(t),

x _i(t)＝x(t)＋n _i?(t)；

In formula, i=1～M, x _i(t) for adding the signal after white Gaussian noise;

6) to x _i(t) gather respectively empirical mode decomposition, obtain K intrinsic mode function component and a remainder r _i(t):

$x_i\left(t\right)=\underset{j=1}{\overset{K}{\mathrm{\&Sigma;}}}{c}_{\mathrm{ij}}\left(t\right)+r_i\left(t\right);$

In formula, c _ij(t) represent to add for the i time after white Gaussian noise, decompose resulting j intrinsic mode function component, j=1～K; x _i(t) represent to add for the i time the signal after white noise, r _i(t) represent to add for the i time white noise by EMD, to decompose the remainder obtaining.

7) intrinsic mode function component corresponding to above step carried out to population mean computing, obtain gathering the intrinsic mode function component c after empirical mode decomposition _jand remainder r (t) (t):

$c_j\left(t\right)=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{c}_{\mathrm{ij}}\left(t\right),$

$r\left(t\right)=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{r}_i\left(t\right);$

In formula, c _j(t) represent original signal to gather resulting j intrinsic mode function component after empirical mode decomposition;

8) obtain intrinsic mode function component and a remainder of K the ascending ordered arrangement of characteristic time scale:

$x\left(t\right)=\underset{j=1}{\overset{K}{\mathrm{\&Sigma;}}}{c}_j\left(t\right)+r\left(t\right).$

9) IMF component being carried out to Hilbert is transformed to:

$H\left[c_i\left(t\right)\right]={\hat{c}}_i\left(t\right)=\frac{1}{\mathrm{\&pi;}}{\&Integral;}_{-\&infin;}^{\&infin;}\frac{c_i\left(t\right)}{t-\mathrm{\&tau;}}\mathrm{d\&tau;}(i=1~n)$

Wherein, H[c _i(t)] represent Hilber t transformation results, represent Hilbert transformation results, c _i(t) represent i IMF component, t represents the time, and τ represents integration amount,

So obtain amplitude function:

$a_i=\sqrt{c_i^2\left(t\right)+{\hat{c}}_i^2\left(t\right)}$

And phase function:

Instantaneous frequency is:

The instantaneous frequency obtaining and instantaneous amplitude combination just can be obtained to Hilbert spectrum:

$H(\mathrm{\&omega;},t)=\mathrm{RP}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{a}_i\left(t\right)e^{j\&Integral;f\left(t\right)\mathrm{dt}}$

In formula, RP represents to get real part.

By the integration of time being got final product to the Hilbert marginal spectrum h (ω) of picked up signal:

$h\left(\mathrm{\&omega;}\right)={\&Integral;}_0^{T}H(\mathrm{\&omega;},t)\mathrm{dt}.$

Below illustrate the present embodiment:

The first step: receive fault-signal to be analyzed.For there is the vibration signal of the rolling bearing of outer ring fault in the original signal x in the present embodiment (t), signal sampling frequency is 12000Hz, and sampling length is 1024 points, and rotating speed is 800 revs/min.Bearing designation is UN205, and rolling body diameter is 7.5mm, and rolling body number is 12, and pitch diameter is 38.7mm, and contact angle is 0 degree.Turning is 29.4Hz frequently, and bearing outer ring failure-frequency is 64.5Hz.The waveform of original signal is as Fig. 2, and its amplitude spectrum as shown in Figure 3, is observed Fig. 3 and failed to find fault signature.

Second step: obtain the standard deviation of original signal x (t), add respectively the white noise n' of 50 different amplitude coefficients _i(t), white noise amplitude coefficient is shown below:

$k\left(i\right)=\frac{i}{100}\mathrm{\&sigma;}(i=1~N);$

The 3rd step: calculate upper and lower extreme point sequence extr_max (x1, y1) and the extr_min (x2, y2) at every turn add signal after white noise, x1, y1(x2, y2) be horizontal stroke, the ordinate of corresponding extreme point;

The 4th step: the amplitude difference standard deviation and poor long-pending std_max and the std_min of separation criteria that calculate extr_max (x1, y1) and extr_min (x2, y2) consecutive point;

The 5th step: calculate and make std_max and std_min reach the white noise amplitude coefficient k_max of minimum value and the mean value k=0.09 σ of k_min, expectation resolution error e=0.01 is set, calculate population mean number of times M=41;

The 6th step: signal x (t) is carried out to EEMD decomposition, the parameter k=0.09 σ of EEMD, M=41.First IMF component obtaining as shown in Figure 4.From Fig. 4, can obviously observe periodic shock, its impulse period T=15.5ms, frequency of impact f=1/T=64.51Hz, matches with the bearing outer ring failure-frequency that calculated above.Calculate the amplitude spectrum of IMF1, as shown in Figure 5, main frequency is 64.52 and the multiple frequence such as 128.91Hz, 199.22Hz, 257.81Hz, coincide with the outer ring fault characteristic frequency calculating, and has extracted accurately housing washer fault characteristic information simultaneously.

The EEMD mechanical fault feature extracting method of the present embodiment based on optimized parameter, by the parameter of itself choosing EEMD right according to signal to be analyzed, then carry out EEMD and decompose the IMF component that obtains characterizing fault signature, ask for its amplitude spectrum, extracted accurately housing washer fault characteristic information.Feasibility of the present invention and validity have been absolutely proved.

The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, obviously, those skilled in the art can carry out various changes and modification and not depart from the spirit and scope of the present invention the present invention.Like this, if within of the present invention these are revised and modification belongs to the scope of the claims in the present invention and equivalent technologies thereof, the present invention is also intended to comprise these changes and modification interior.

Claims (7)

1. the mechanical fault feature extracting method based on optimized parameter set empirical mode decomposition, is characterized in that: comprise the steps:

1) input signal x (t), calculating adds the product std_max of the maximum value sequence consecutive point amplitude difference of signal and the standard deviation of spacing after the white noise of different amplitude coefficients, and the product std_min of the standard deviation of minimal value sequence consecutive point amplitude difference and spacing, acquisition makes std_max obtain the white noise amplitude coefficient k_max of minimum value, obtains and makes std_min obtain the white noise amplitude coefficient k_min of minimum value;

2) ask the mean value of k_max and k_min as the parameter k of EEMD, then according to the expectation resolution error e calculated population average time M arranging, complete EEMD parameter optimization;

3) fault-signal is carried out to EEMD decomposition, obtain a series of intrinsic mode function IMF components, complete fault signature and extract.

2. the mechanical fault feature extracting method based on optimized parameter set empirical mode decomposition according to claim 1, is characterized in that: described step 1) specifically comprise the steps:

11) obtain the standard deviation of original signal x (t), add respectively the white noise n of N different amplitude coefficients ^' _i(t), white noise amplitude coefficient is shown below:

$k\left(i\right)=\frac{i}{100}\mathrm{\&sigma;}(i=1~N),$

Wherein, k (i) represents white noise amplitude coefficient, and i represents the i time and adds white noise;

12) calculate the upper extreme point sequence extr_max (x1 at every turn add signal after white noise, y1), calculate the lower extreme point sequence extr_min (x2 at every turn add signal after white noise, y2), wherein, (x1, y1), (x2, y2) are horizontal stroke, the ordinate of corresponding extreme point;

13) calculate amplitude difference standard deviation and the poor long-pending std_max of separation criteria of extr_max (x1, y1) consecutive point, calculate amplitude difference standard deviation and the poor long-pending std_min of separation criteria of extr_min (x2, y2) consecutive point, as shown in the formula:

$\mathrm{std}\_\max =\sqrt{\frac{1}{N1-1}\underset{i=N1}{\overset{N1-1}{\mathrm{\&Sigma;}}}\left(\right|{x1}_{i+1}-{x1}_i|-u_1)}\&times;\sqrt{\frac{1}{N1-1}\underset{i=N1}{\overset{N1-1}{\mathrm{\&Sigma;}}}\left(\right|{y1}_{i+1}-{y1}_i|-u_1^{\&prime;})}$

$\mathrm{std}\_\min =\sqrt{\frac{1}{N2-1}\underset{i=N2}{\overset{N2-1}{\mathrm{\&Sigma;}}}\left(\right|{x2}_{i+1}-{x2}_i|-u_2)}\&times;\sqrt{\frac{1}{N2-1}\underset{i=N2}{\overset{N2-1}{\mathrm{\&Sigma;}}}\left(\right|{y2}_{i+1}-{y2}_i|-u_2^{\&prime;})}$

In formula, N1 is counting of extr_max (x1, y1), and N2 is counting of extr_min (x2, y2), u ₁, for the mean value of extr_max (x1, y1) consecutive point amplitude difference and the mean value of spacing, u ₂, for the mean value of extr_min (x2, y2) consecutive point amplitude difference and the mean value of spacing, x1 _i+1represent maximum point horizontal ordinate, x1 _irepresent maximum point horizontal ordinate, y1 _i+1represent maximum point ordinate, y1 _irepresent maximum point ordinate, x2 _i+1represent minimum point horizontal ordinate, x2 _irepresent minimum point horizontal ordinate, y2 _i+1represent minimum point ordinate, y2 _irepresent minimum point ordinate.

3. the mechanical fault feature extracting method based on optimized parameter set empirical mode decomposition according to claim 1, is characterized in that: described step 2) specifically comprise the steps:

14) calculate and make std_max and std_min reach the white noise amplitude coefficient k_max of minimum value and the mean value k of k_min, then calculate M according to following formula:

$M={\left(\frac{k}{e}\right)}^2,$

And whether the size that judges M be less than 20, if during M < 20, get M=20.

4. the mechanical fault feature extracting method based on optimized parameter set empirical mode decomposition according to claim 1, is characterized in that: described step 3) specifically comprise the steps:

15) in original vibration signal x (t), adding respectively in such a way M average is 0, the white Gaussian noise n that amplitude coefficient is k _i(t):

x _i(t)＝x(t)+n _i(t)，

In formula, i=1～M, i represents the i time and adds white noise, x _i(t) for adding the signal after white Gaussian noise;

16) to x _i(t) gather respectively empirical mode decomposition, obtain K intrinsic mode function component and a remainder r _i(t):

$x_i\left(t\right)=\underset{j=1}{\overset{K}{\mathrm{\&Sigma;}}}{c}_{\mathrm{ij}}\left(t\right)+r_i\left(t\right)$

In formula, c _ij(t) represent to add for the i time after white Gaussian noise, decompose resulting j intrinsic mode function component, j=1～K; r _i(t) represent to add for the i time the set empirical mode decomposition remainder of white Gaussian noise signal afterwards;

17) by above-mentioned steps 16) corresponding intrinsic mode function component carries out population mean computing, obtains gathering the intrinsic mode function component c after empirical mode decomposition _jand remainder r (t) (t):

$c_j\left(t\right)=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{c}_{\mathrm{ij}}\left(t\right),$

$r\left(t\right)=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{r}_i\left(t\right),$

In formula, c _j(t) represent original signal to gather resulting j intrinsic mode function component after empirical mode decomposition;

18) obtain intrinsic mode function component and a remainder of K the ascending ordered arrangement of characteristic time scale:

$x\left(t\right)=\underset{j=1}{\overset{K}{\mathrm{\&Sigma;}}}{c}_j\left(t\right)+r\left(t\right),$

Wherein, x (t) represents original signal, c _j(t) represent j IMF component, r (t) represents remainder.

5. the mechanical fault feature extracting method based on optimized parameter set empirical mode decomposition according to claim 4, is characterized in that: the fault signature described step 3) is that the amplitude spectrum of the IMF component after decomposing by EEMD and Hilbert have composed fault signature and extract.

6. the mechanical fault feature extracting method based on optimized parameter set empirical mode decomposition according to claim 1, is characterized in that: described error e gets 0.01.

7. the mechanical fault feature extracting method based on optimized parameter set empirical mode decomposition according to claim 2, is characterized in that: described different amplitude coefficient white noise n ^' _i(t) the times N that adds gets 50.