A METHOD AND AN APPARATUS FOR PREDICTING THE CONDITION OF A MACHINE OR A COMPONENT OF THE MACHINE

• US 20150377745A1
• Filed: 01/20/2014
• Published: 12/31/2015
• Est. Priority Date: 02/19/2013
• Status: Active Grant

• Current Assignee: E-Maintenance Sweden AB
• Original Assignee: E-Maintenance Sweden AB
• Inventors: Al-Najjar, Basim
• Granted Patent: US 9,476,803 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01M 99/005 Testing of complete machines, e.g. washing-machines or mobile phones testing of machine parts G01M13/00; testing of electric apparatus or components G01R31/50
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: G05B 23/0283 Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06
• CPC Class Codes: G07C 3/00 Registering or indicating the condition or the working of machines or other apparatus, other than vehicles

Abstract

The present invention relates to a method and an apparatus for predicting the condition of a machine or a component of the machine based on measurements of a condition monitoring parameter. The apparatus comprising—a data storage (6) for storing a moving window including measurement values of the condition monitoring parameter,—a monitoring module (7) configured to determine when a measurement value exceeds a potential failure level at which damage of the machine or component is considered initiated, and to determine the point in time (t_P) when the monitoring parameter exceeded the potential failure level,—a data cleaning module (8) configured to discard a new measurement value if it is smaller than the latest stored measurement value in the moving window, to determine an expected value for the measurement based on a defined pattern for the condition monitoring parameter, which pattern reflects an expected behavior of the monitoring parameter after damage has been initiated, and the time elapsed since the point in time the monitoring parameter exceeded the potential failure level, and to discard the new measurement value if it deviates too much from the expected value, and to store not discarded measurement values in the moving window, and—a prediction module (10) configured to predict the level of the condition monitoring parameter at a future point in time based on the measurement values stored in the moving window.

Claims

1. A method for predicting the condition of a machine or a component of the machine based on measurements of a condition monitoring parameter, wherein a potential failure level of the condition monitoring parameter has been defined, at which level damage of the machine or component is considered initiated, wherein

a pattern for the condition monitoring parameter has been defined, which pattern reflects an expected behavior of the monitoring parameter after damage has been initiated, and a plurality of measurement values of the condition monitoring parameter have been stored in a moving window, and

the method comprises the steps of:

a) obtaining a new measurement value of the condition monitoring parameter,

b) comparing the new measurement value with the latest stored measurement value, and discard the new measurement value if it is smaller than the latest stored measurement value, and otherwise replace the oldest of the measurement values stored in the moving window with the new measurement value,

c) comparing the new measurement value with the potential failure level of the condition monitoring parameter, and repeating the steps a-c until the new measurement value exceeds the potential failure level of the condition monitoring parameter,

d) determining the point in time when the monitoring parameter exceeded the potential failure level,

e) receiving a new measurement value of the condition monitoring parameter,

f) comparing the new measurement value with the latest stored measurement value, and discarded the new measurement value if it is smaller than the latest stored measurement value,

g) comparing the new measurement value with an expected value determined in accordance with said defined pattern for the condition monitoring parameter and the time elapsed since the point in time the monitoring parameter exceeded the potential failure level, and discard the new measurement value if it deviates too much from the pattern, and otherwise replace the oldest of the measurement values stored in the moving window with the new measurement value, and

h) predicting the level of the condition monitoring parameter at a future point in time based on the measurement values stored in the moving window.

2. The method according to claim 1, wherein the moving window includes at least three stored measurement values.

3. The method according to claim 1, wherein said defined pattern for the condition monitoring parameter is a curve which starts increasing at the point in time the monitoring parameter exceeded the potential failure level.

4. The method according to claim 1, wherein the new measurement value deviates too much from the expected value if the measurement deviates from the expected value with more than a limit value.

5. The method according to claim 1, wherein if the measurement deviates too much from the expected value, the step g further includes:

generating a message to a user to control the measurement and provide a new measurement,

receiving a new measurement value,

comparing the new measurement value with an expected value determined in accordance with said defined pattern for the condition monitoring parameter and the time elapsed since the point in time the monitoring parameter exceeded the potential failure level, and

generating a warning if the new measurement value also deviates too much from the expected value, and otherwise replace the oldest measurement value stored in the moving window with the new measurement value.

6. The method according to claim 1, wherein the method further comprises the steps of:

i) comparing the predicted level of the condition monitoring parameter with at least one limit level for the condition monitoring parameter, repeating steps e-i at least until the predicted level of the condition monitoring parameter exceeds the limit level, and

j) generating a message regarding a maintenance action when the predicted level of the condition monitoring parameter exceeds said limit level.

7. The method according to claim 1, wherein the machine is a rotating machine and the condition monitoring measurement values are any of vibration measurements, temperature measurements, sound measurements, and shock pulse measurements.

10. A computer-readable medium having a program recorded thereon, where the program is to make a computer perform the steps of claim 1, when said program is run on the computer.

8. An apparatus for predicting the condition of a machine or a component of the machine based on measurements of a condition monitoring parameter, comprising:

a communication unit configured to obtain measurement values of the condition monitoring parameter,

a data storage for storing a moving window including measurement values of the condition monitoring parameter,

monitoring module configured to monitor the measurement values and to determine when a measurement value exceeds a potential failure level at which damage of the machine or component is considered initiated, and to determine the point in time when the monitoring parameter exceeded the potential failure level,

a data cleaning module configured to discard a new measurement value if it is smaller than the latest stored measurement value in the moving window, to determine an expected value for the measurement based on a defined pattern for the condition monitoring parameter, which pattern reflects an expected behavior of the monitoring parameter after damage has been initiated, and the time elapsed since the point in time the monitoring parameter exceeded the potential failure level, to discard the new measurement value if it deviates too much from the expected value, to store not discarded measurement values in the moving window, and to replace the oldest of the measurement values stored in the moving window with the new measurement value if the moving window is full, and

a prediction module configured to predict the level of the condition monitoring parameter at a future point in time based on the measurement values stored in the moving window.

9. The apparatus according to claim 8, wherein the apparatus further comprises a human machine interface configured to provide messages regarding maintenance actions to a user, and the monitoring module is configured to compare the predicted level of the condition monitoring parameter with at least one limit level for the condition monitoring parameter, and to generate a message regarding a maintenance action when the predicted level of the condition monitoring parameter exceeds said limit level.