Method and sensor arrangement for load measurement on rolling element bearing based on model deformation

• US 20070143039A1
• Filed: 09/16/2004
• Published: 06/21/2007
• Est. Priority Date: 09/24/2003
• Status: Active Grant

• Current Assignee: AB SKF
• Original Assignee: AB SKF
• Inventors: Mol, Hendrik
• Granted Patent: US 7,389,701 B2
• Time in Patent Office: Days
• US Class Current: 702/41
• CPC Class Codes: F16C 19/522 related to load on the bearing, e.g. bearings with load sensors or means to protect the bearing against overload
• CPC Class Codes: F16C 2233/00 Monitoring condition, e.g. temperature, load, vibration
• CPC Class Codes: G01L 5/0009 Force sensors associated with a bearing testing of bearings G01M13/04
• CPC Class Codes: G01M 13/045 Acoustic or vibration analysis

Abstract

Method and sensor arrangement for determining a load vector acting on a rolling element bearing (1) in operation. A plurality of N sensors (8) are provided which measure displacement and/or strain for determining displacement and/or strain in one of the elements (5, 6, 7) of the rolling element bearing (1). Furthermore, a mode shape coefficients calculator (11) is provided, connected to the plurality of N sensors (8), for determining a deformation of the element (5, 6, 7) by calculating amplitude and phase of N/2 Fourier terms representing at least one radial mode shape of the ring shape element (5, 6, 7). Also, a bearing neural network (12) is present, connected to the mode shape coefficients calculator (11), the bearing neural network (12) being trained to provide the load vector on the rolling element bearing (1) from the N/2 Fourier terms.

Claims

1. Method for determining a load vector acting on a rolling element bearing, in which the method comprises:

measuring displacement and/or strain using N sensors for determining displacement and/or strain in a ring shaped elements (of the rolling element bearing;

determining a deformation of the element by calculating amplitude and phase of N/2 Fourier terms representing at least one radial mode shape of the ring shaped element;

feeding the N/2 Fourier terms to a bearing neural network, the bearing neural network being trained to provide the load vector on the rolling element bearing from the N/2 Fourier terms.

2. Method according to claim 1, in which the rolling element bearing comprises two rows of coaxial bearings, the ring shaped element of the rolling element bearing being the bearing outer ring of one of the two rows of coaxial bearings.

3. Method according to claim 1, in which the N sensors comprise strain sensors, and the at least one radial mode shape comprises a mode shape of order zero and one or more mode shapes of order two or higher.

4. Method according to claim 1, in which the N sensors comprise displacement sensors and the least one radial mode shape comprises one or more mode shapes of order zero and higher.

5. Method according to claim 1, further comprising determining N/2 Chebyshev polynomial coefficients from the N sensor signals, representing at least one axial mode shape of the ring shaped element.

6. Method according to claim 1, in which the bearing neural network is trained using a plurality of data sets, each data set comprising a predefined load vector on a specific type of rolling element bearing and associated measurement data from the N sensors.

14. Method according to claim 2, further comprising determining N/2 Chebyshev polynomial coefficients from the N sensor signals, representing at least one axial mode shape of the ring shaped element.

15. Method according to claim 2, in which the bearing neural network is trained using a plurality of data sets, each data set comprising a predefined load vector on a specific type of rolling element bearing and associated measurement data from the N sensors.

7. Sensor arrangement for determining a load vector acting on a rolling element bearing in operation, the sensor arrangement comprising a plurality of N sensors which measure displacement and/or strain for determining displacement and/or strain in one of the elements of the rolling element bearing;

a mode shape coefficients calculator, connected to the plurality of N sensors, for determining a deformation of the element by calculating amplitude and phase of N/2 Fourier terms representing at least one radial mode shape of the ring shaped element;

a bearing neural network, connected to the mode shape coefficients calculator, the bearing neural network being trained to provide the load vector on the rolling element bearing from the N/2 Fourier terms.

8. Sensor arrangement according to claim 7, in which the rolling element bearing comprises two rows of coaxial bearings, the ring shaped element of the rolling element bearing being the bearing outer ring of one of the two rows of coaxial bearings.

9. Sensor arrangement according to claim 7, in which the N sensors comprise strain sensors, and the at least one radial mode shape comprises a mode shape of order zero and one or more mode shapes of order two or higher.

10. Sensor arrangement according to claim 7, in which the N sensors comprise displacement sensors, and the at least one radial mode shape comprises one or more mode shapes of order zero or higher.

11. Sensor arrangement according to claim 7, in which the mode shape coefficient calculator is further arranged to determine N/2 Chebyshev polynomial coefficients representing at least one axial mode shape of the ring shaped element.

12. Sensor arrangement according to claim 7, in which the bearing neural network is trained using a plurality of data sets, each data set comprising a predefined load vector on a specific type of rolling element bearing and associated measurement data from the N sensors.

13. Sensor arrangement according to claim 7, in which the bearing inner ring or outer ring are attached to a sensor holder, a circumferential recession being provided between at least part of the contacting surfaces of the inner ring or outer ring and the sensor holder.

16. Sensor arrangement according to claim 8, in which the N sensors comprise strain sensors, and the at least one radial mode shape comprises a mode shape or order zero and one or more mode shapes of order two or higher.

17. Sensor arrangement according to claim 8, in which the N sensors comprise displacement sensors, and the at least one radial mode shape comprises one or more mode shapes of order zero or higher.

18. Sensor arrangement according to claim 8, in which the bearing inner ring or outer ring are attached to a sensor holder, a circumferential recession being provided between at least part of the contacting surfaces of the inner ring or outer ring and the sensor holder.

19. Sensor arrangement according to claim 9, in which the bearing inner ring or outer ring are attached to a sensor holder, a circumferential recession being provided between at least part of the contacting surfaces of the inner ring or outer ring and the sensor holder.

20. Sensor arrangement according to claim 10, in which the bearing inner ring or outer ring are attached to a sensor holder, a circumferential recession being provided between at least part of the contacting surfaces of the inner ring or outer ring and the sensor holder.