Method and data processing device for severity assessment of bearing defects using vibration energy

• US 9,989,439 B2
• Filed: 10/14/2016
• Issued: 06/05/2018
• Est. Priority Date: 10/20/2015
• Status: Active Grant

• Current Assignee: ABet SKF
• Original Assignee: ABet SKF
• Inventors: Thomson, Allan
• Primary Examiner(s): DO, AN H
• Application Number:
  US15/294,064
• Publication Number:
  US 20170108406A1
• Time in Patent Office: 599 Days
• Field of Search: 702 33, 702 35, 702 39, 702 56, 73659, 73660
• CPC Class Codes: G01H 1/00 Measuring characteristics of vibrations in solids by using direct conduction to the detector G01H9/00, G01H11/00 take precedence
• CPC Class Codes: G01M 13/045 Acoustic or vibration analysis
• CPC Class Codes: G01N 29/4463 Signal correction, e.g. distance amplitude correction [DAC], distance gain size [DGS], noise filtering

Abstract

Method for detecting bearing defect severity based on bearing rotation speed and at least one data sample of sensor data obtained to measure vibrations of a bearing ring is provided. The method includes converting the data sample from time domain to frequency domain to obtain a signal frequency spectrum; determining a defect center frequency of the bearing using the rotation speed; and identifying a predetermined number of frequency peaks of the signal frequency spectrum. A total vibration energy in an overall frequency band including the predetermined number of frequency peaks is first determined; next, for each of the frequency peaks, a peak energy as a spectral energy of signal components giving rise to the frequency peaks is determined; calculating a bearing defect spectral energy using the peak energies; and finally a ratio of the bearing defect spectral energy and a total vibration energy to assess a defect severity is obtained.

Claims

1. A method for detecting a bearing defect severity based on a rotation speed of a bearing and on at least one data sample of sensor data obtained by a sensor configured to measure vibrations of a bearing ring, the method comprises the steps of:

a. converting the data sample from time domain to frequency domain by applying a Fourier Transform to obtain a signal frequency spectrum;

b. determining a defect center frequency of the bearing using the rotation speed;

c. identifying a predetermined number of frequency peaks of the signal frequency spectrum;

d. determining a total vibration energy in an overall frequency band including the predetermined number of frequency peaks;

e. determining, for each of the frequency peaks, a peak energy as a spectral energy of signal components giving rise to the frequency peaks;

f. calculating a bearing defect spectral energy using the peak energies of the frequency peaks; and

g. using a ratio of the bearing defect spectral energy and the total vibration energy to assess a severity of the bearing defect.

2. The method according to claim 1, wherein the step of identifying the predetermined number of frequency peaks includes searching a first frequency peak in a frequency band including the defect center frequency and searching second and subsequent frequency peaks in frequency bands including integer multiples of the defect center frequency.

3. The method according to claim 1, wherein the step of identifying the predetermined number of frequency peaks includes the steps of:

a. determining a predetermined number of frequency bands (W1-W5) of a first width using the defect center frequency , wherein a first frequency band includes the defect center frequency and the second and following frequency bands include integer multiples the defect center frequency as their respective window center frequency;

b. determining, for each of the frequency bands (W1-W5), a local maximum of the signal frequency spectrum within the respective frequency band (W1-W5); and

c. determining peak energies of the signal frequency spectrum within each of the frequency bands (W1-W5) by calculating a squared sum of an amplitude of the signal frequency spectrum at the local maximum and of the amplitudes of the signal frequency spectrum adjacent to the local maximum.

4. The method according to claim 1, wherein the step of calculating a bearing defect spectral energy includes of calculating a sum, weighted sum or root sum square of the peak energies.

5. The method according to claim 1, wherein the step of determining a total vibration energy in the overall frequency band includes calculating the root squared sum of the magnitudes of the signal frequency spectrum within the frequency band.

6. The method according to claim 1, wherein the step of upper limit and the lower limit of the overall frequency band are application dependent settings read from a memory device.

7. The method according to claim 1, further comprising the steps of:

a. comparing the ratio with at least two threshold values to classify the defect severity into at least three severity classes, and

b. outputting the result.

8. The method according to claim 7, wherein the at least two threshold values are application-dependent settings read from a storage device.

9. A condition monitoring system for monitoring a machine comprising:

at least one bearing equipped with a sensor configured to measure vibrations of a bearing ring, and

a data processing device, wherein the data processing device is configured to process data samples of sensor data obtained by the sensor by converting the data sample from time domain to frequency domain by applying a Fourier Transform to obtain a signal frequency spectrum;

determining a defect center frequency of the bearing using the rotation speed;

identifying a predetermined number of frequency peaks of the signal frequency spectrum;

determining a total vibration energy in an overall frequency band including the predetermined number of frequency peaks;

determining, for each of the frequency peaks, a peak energy as a spectral energy of signal components giving rise to the frequency peaks;

calculating a bearing defect spectral energy using the peak energies of the frequency peaks; and

using a ratio of the bearing defect spectral energy and the total vibration energy to assess a severity of the bearing defect.

10. A machine comprising:

at least one bearing equipped with a sensor, and

a condition monitoring system having at least one bearing equipped with a sensor configured to measure vibrations of a bearing ring, and

a data processing device, wherein the data processing device is configured to process data samples of sensor data obtained by the sensor by converting the data sample from time domain to frequency domain by applying a Fourier Transform to obtain a signal frequency spectrum;

determining a defect center frequency of the bearing using the rotation speed;

identifying a predetermined number of frequency peaks of the signal frequency spectrum;

determining a total vibration energy in an overall frequency band including the predetermined number of frequency peaks;

determining, for each of the frequency peaks, a peak energy as a spectral energy of signal components giving rise to the frequency peaks;

calculating a bearing defect spectral energy using the peak energies of the frequency peaks; and

using a ratio of the bearing defect spectral energy and the total vibration energy to assess a severity of the bearing defect, wherein

the data processing device of the condition monitoring system is configured to process data samples of sensor data obtained by the sensor.