Intelligence identification method for vibration characteristic of rotating machinery

1. Intelligent identification method for a vibration characteristic of rotating machinery, comprising steps in the following order:

step 1, converting a speed or acceleration time domain signal of mechanical vibration to a frequency domain envelope spectrum by signal processing, and recording a frequency upper limit value f max of the envelope spectrum;

step 2, at least screening out a high energy harmonic with a frequency range within f max /N max by amplitude comparison, wherein N max is a frequency multiple upper limit multiple for performing a frequency multiple check on the high energy harmonic;

step 3, extracting at least one set of characteristic parameters, based on respective amplitudes and/or frequencies, of 1-fold to N max -fold frequency region peaks of each high energy harmonic, wherein the 1-fold frequency region peak is the high energy harmonic itself; and

step 4, inputting the at least one set of characteristic parameters of each high energy harmonic into a machine learning intelligent algorithm to perform training and calculation.

2. Intelligent identification method according to claim 1 , wherein the at least one set of characteristic parameters comprises at least one of the following three sets of characteristic parameters: a set of absolute amplitude data, a set of relative amplitude data and a set of frequency derived data of the 1-fold to N max -fold frequency region peaks of each high energy harmonic.

3. Intelligent identification method according to claim 2 , wherein the set of relative amplitude data is a set of bamboo grass ratio data ((BambooGrassRatio y 1 ), BambooGrassRatio (y 2 ) . . . BambooGrassRatio(y N max )) or a set of bamboo grass ratio coefficient data (S 1 , S 2 . . . S N max ), and the set of frequency derived data is a set of frequency multiple deviation degree data (dev(f 1 ), dev (f 2 ) . . . dev(f N max )).

4. Intelligent identification method according to claim 1 , wherein the amplitude comparison in step 2 comprises sub-steps in the following order:

sub-step A, screening of local peaks;

sub-step B, correction of local peaks, and replacing original values with corrected values; and

sub-step C, bamboo grass filtering.

5. Intelligent identification method according to claim 4 , wherein the bamboo grass filtering uses a triangular model or normal distribution model to distribute weight coefficients of comparison samples in a nearby window.

6. Intelligent identification method according to claim 1 , wherein the value of the frequency multiple upper limit multiple N max is in the range of 3 to 10.

7. Intelligent identification method according to claim 6 , wherein the value of the frequency multiple upper limit multiple N max is in the range of 4 to 7.

8. Intelligent identification method according to claim 7 , wherein the value of the frequency multiple upper limit multiple N max is in the range of 5 to 6.

9. Intelligent identification method according to claim 1 , wherein the machine learning intelligent algorithm is a fully connected neural network, comprising a layer structure distribution and an input unit quantity adapted to the quantity of the at least one set of characteristic parameters.

10. Intelligent identification method according to claim 1 , further comprising a step 5: checking whether a frequency of the high energy harmonic is the same as a defect characteristic frequency of the rotating machinery.