Method for determining a remaining useful life of a bearing

Patent metadata

Field	Value
Publication number	US-11885708-B2
Application number	US-202017794953-A
Country	US
Kind code	B2
Filing date	Nov 27, 2020
Priority date	Jan 23, 2020
Publication date	Jan 30, 2024
Grant date	Jan 30, 2024

Abstract

A method for determining a remaining useful life of a bearing having a surface defect includes determining a defect size of the surface defect based on oscillations of the bearing. The method also includes determining, based on the defect size, at least one of the principal stresses or a contact force of the bearing caused by the surface defect. The method also includes determining the remaining useful life of the bearing based on at least one of the principal stresses and/or the contact force.

First claim

The invention claimed is:

1. A method for determining a remaining useful life of a bearing having a surface defect, the method comprising:
determining a defect size of the surface defect based on sampled oscillations of the bearing, wherein determining the defect size of the surface defect comprises determining the defect size of the surface defect based on associating, using a trained machine learning model, the sampled oscillations of the bearing with the defect size of the surface defect;
determining, based on the defect size, at least one of principal stresses or a contact force of the bearing caused by the surface defect, wherein the determining of the at least one of the principal stresses or the contact force is based on a substitute model of the bearing, wherein the defect size of the surface defect is an input of the substitute model, and wherein the substitute model associates the defect size with a principal stress of the at least one of the principal stresses or the contact force; and
determining the remaining useful life of the bearing based on at least one of the principal stresses, the contact force, or the at least one of the principal stresses and the contact force.

2. The method of claim 1 , wherein the determining of the remaining useful life of the bearing is based on a maximum principal stress.

3. The method of claim 1 , wherein the determining of the remaining useful life of the bearing is based on a high-cycle-fatigue model.

4. The method of claim 1 , further comprising:
loading the substitute model of the bearing,
wherein one or more parameters of the substitute model are based on operational conditions of the bearing, model specific bearing data, or a combination thereof.

5. The method of claim 1 , wherein the remaining useful life is determined based on a high-cycle-fatigue model,
wherein the at least one of the principal stresses is an input for the high-cycle-fatigue model, and
wherein a cycle count and material degradation are parameters of the high-cycle-fatigue model.

6. The method of claim 1 , wherein determining the defect size of the surface defect based on the sampled oscillations of the bearing comprises determining the defect size of the surface defect based on sampled harmonic oscillations of an inner ring, an outer ring, or the inner ring and the outer ring of the bearing.

7. A system configured to determine a remaining useful life of a bearing having a surface defect, the system comprising:
an apparatus comprising:
a processor,
a trained machine learning model and a substitute model of the bearing; and
a sensor operable to measure oscillations of a bearing,

wherein the processor is configured to:
determine a defect size of the surface defect based on sampled oscillations of the bearing, wherein the determination of the defect size of the surface defect comprises determination of the defect size of the surface defect based on association, using the trained machine learning model, the sampled oscillations of the bearing with the defect size of the surface defect;
determine, based on the defect size, at least one of principal stresses or a contact force of the bearing caused by the surface defect, wherein the determination of the at least one of the principal stresses or the contact force is based on the substitute model of the bearing, wherein the defect size of the surface defect is an input of the substitute model, and wherein the substitute model associates the defect size with a principal stress of the at least one of the principal stresses or the contact force; and
determine the remaining useful life of the bearing based on at least one of the principal stresses, the contact force, or the at least one of the principal stresses and the contact force.

8. In a non-transitory computer-readable storage medium that stores instructions executable by one or more processors to determine a remaining useful life of a bearing having a surface defect, the instructions comprising:
determining a defect size of the surface defect based on sampled oscillations of the bearing, wherein determining the defect size of the surface defect comprises determining the defect size of the surface defect based on associating, using a trained machine learning model, the sampled oscillations of the bearing with a defect size of the surface defect;
determining, based on the defect size, at least one of principal stresses or a contact force of the bearing caused by the surface defect, wherein the determining of the at least one of the principal stresses or the contact force is based on a substitute model of the bearing, wherein the defect size of the surface defect is an input of the substitute model, and wherein the substitute model associates the defect size with a principal stress of the at least one of the principal stresses or the contact force; and
determining the remaining useful life of the bearing based on at least one of the principal stresses, the contact force, or the at least one of the principal stresses and the contact force.

9. The non-transitory computer-readable storage medium of claim 8 , wherein the determining of the remaining useful life of the bearing is based on a maximum principal stress.

10. The non-transitory computer-readable storage medium of claim 8 , wherein the determining of the remaining useful life of the bearing is based on a high-cycle-fatigue model.

11. The non-transitory computer-readable storage medium of claim 8 , wherein the instructions further comprise:
loading the substitute model of the bearing,
wherein one or more parameters of the substitute model are based on operational conditions of the bearing, model specific bearing data, or a combination thereof.

12. The non-transitory computer-readable storage medium of claim 8 , wherein the remaining useful life is determined based on a high-cycle-fatigue model,
wherein the at least one of the principal stresses is an input for the high-cycle-fatigue model, and
wherein a cycle count and material degradation are parameters of the high-cycle-fatigue model.

Assignees

• Siemens Ag

Inventors

• Erlwein Matthias
• Malik Vincent
• Wolf Pozzo Christian Andreas

Classifications

• G01M13/045 Primary

  Acoustic or vibration analysis · CPC title