System, apparatus, and method of determining condition of a bearing

Patent metadata

Field	Value
Publication number	US-12320384-B2
Application number	US-202017622707-A
Country	US
Kind code	B2
Filing date	Jun 26, 2020
Priority date	Jun 26, 2019
Publication date	Jun 3, 2025
Grant date	Jun 3, 2025

Abstract

A system, apparatus, and method of determining a condition of at least one bearing in a system are provided. The method includes receiving operation data associated with the system from one or more sensing units associated with the system and determining an operation profile of the at least one bearing from the operation data. The operation profile includes a vibration response, a thermal response, and/or a frequency response associated with the at least one bearing. An impact force profile is determined during operation of the at least one bearing based on the operation profile and a virtual bearing model trained on operation profiles and impact force profiles associated with a group of bearings comparable with the at least one bearing. The condition of the at least one bearing is determined based on the impact force profile.

First claim

The invention claimed is:

1. A method of determining a condition of at least one bearing in a system, the method being computer-implemented and comprising:
receiving operation data associated with the system from one or more sensors associated with the system, the one or more sensors comprising a vibration sensor, a current sensor, a voltage sensor, a temperature sensor, a magnetic flux sensor, a velocity sensor, or any combination thereof;
determining an operation profile of the at least one bearing from the operation data, wherein the operation profile includes a vibration response, a thermal response, or the vibration response and the thermal response associated with the at least one bearing;
determining an impact force profile during operation of the at least one bearing based on the operation profile and a virtual bearing model, wherein the impact force profile is associated with an impact between a rolling element of the at least one bearing and a physical defect of the bearing, wherein the virtual bearing model is trained on operation profiles and impact force profiles associated with a group of bearings comparable with the at least one bearing, and wherein the virtual bearing model is generated based on predetermined defects effected on one or more bearings of the group of bearings by a shaping device; and
determining the condition of the at least one bearing based on the impact force profile, wherein determining the condition of the at least one bearing based on the impact force profile comprises:
identifying a defect in the at least one bearing based on the impact force profile; and
determining a contamination condition of a lubricant in the at least one bearing,
wherein identifying the defect and determining the contamination condition comprise superimposing the operating profile, which is determined from the operation data received from the one or more sensors, of the at least one bearing on the operation profiles in the virtual bearing model.

2. The method of claim 1 , further comprising training the virtual bearing model on the operation profiles and the impact force profiles associated with the group of bearings comparable with the at least one bearing, the training comprising:
determining test operation profiles based on test operation data associated with the group of bearings,
wherein a test operation comprises operation data generated from the group of bearings during testing of the bearings, and
wherein the test operation profiles comprise vibration responses and thermal responses associated with the group of bearings.

3. The method of claim 2 , wherein determining the test operation profiles based on the test operation data associated with the group of bearings comprises:
operating systems including the bearings in one or more system load conditions, wherein the system load conditions indicate a system load on the systems; and
determining the test operation profiles associated with the group of bearings for the system load conditions, wherein the test operation profiles are generated based on the test operation data received from sensors of the one or more sensors positioned within and outside each bearing of the group of bearings.

4. The method of claim 3 , further comprising:
determining a simulated impact force from the test operation profiles and simulated operation profiles; and
generating stress distribution associated with the group of bearings based on the simulated impact force.

5. The method of claim 4 , wherein determining the simulated impact force from the test operation profiles and the simulated operation profiles comprises:
predicting the simulated impact force based on the simulated operation profile and a mass of bearing ball, a damping co-efficient, a stiffness associated with the bearing, or any combination thereof,
wherein the simulated impact force includes a steady component from steady-state movement of the bearings, a dynamic component associated with one or more of the predetermined defects, or the steady component and the dynamic component.

6. The method of claim 4 , wherein predicting the stress distribution associated with the group of bearings based on the simulated impact force further comprises:
comparing the test operation profiles and the simulated operation profiles;
updating the simulated impact force based on the comparison;
generating the stress distribution based on the updated simulated impact force; and
mapping the stress distribution to one or more of the predetermined defects using at least one machine learning algorithm.

7. The method of claim 6 , wherein updating the simulated impact force based on the comparison between the test operation profiles and the simulation operation profiles comprises:
calibrating the updated simulated impact force associated with the group of bearings based on a difference between the test operation profiles and the simulated operation profiles using at least one machine learning algorithm.

8. The method of claim 7 , wherein the at least one machine learning algorithm is a differential evolutionary algorithm, and
wherein calibrating the updated simulated impact force associated with the group of bearings based on the difference between the test operation profiles and the simulated operation profiles using at least one machine learning algorithm comprises:
defining an upper stress limit and a lower stress limit for the group of bearings;
determining a probable stress distribution within the upper stress limit and the lower stress limit through a mutation or a recombination operation, wherein the probable stress distribution is determined based on the difference; and
selecting the stress distribution from the probable stress distribution using continuous function optimization based on impact force difference.

9. The method of claim 2 , wherein determining the test operation profiles based on the test operation data associated the group of bearings comprises:
predicting a life of the bearings when subject to one or more of the predetermined defects based on bearing load, load zone, bearing clearance, lubrication viscosity, lubricant contamination, or any combination thereof associated with the one or more bearings in the group of bearings.

10. The method of claim 9 , further comprising:
predicting a stress distribution associated with the at least one bearing during operation of the at least one bearing, wherein the stress distribution is predicted based on the impact force profile of the at least one bearing and the virtual bearing model; and
predicting a remaining life of the at least one bearing based on the stress distribution and the predicted life using a neural network,
wherein the neural network is configured to perform gradient descent optimization.

11. The method of claim 1 , wherein determining the condition of the at least one bearing based on the impact force profile comprises:
determining fatigue of the at least one bearing with respect to lubricant temperature rise, foreign particles in the lubricant, reduction in oil film parameter of the lubricant, or any combination thereof based on the virtual bearing model.

12. An apparatus for determining a condition of at least one bearing in a system, the apparatus comprising:
one or more processing units; and
a memory unit communicatively coupled to the one or more processing units, wherein the memory unit comprises a bearing module and a virtual bearing module stored in the form of machine-readable instructions executable by the one or more processing units, wherein the bearing module is configured to:
receive operation data associated with the system from one or more sensors associated

Assignees

• Siemens Ag

Inventors

• Nair Sudev
• Malik Vincent

Classifications

• G01M13/045 Primary

  Acoustic or vibration analysis · CPC title

• F16C2233/00

  Monitoring condition, e.g. temperature, load, vibration · CPC title

• F16C19/525

  related to temperature and heat, e.g. insulation · CPC title

• F16C19/527 Primary

  related to vibration and noise · CPC title

• G06F2119/04

  Ageing analysis or optimisation against ageing · CPC title