US 11,746,752 B2
Method and system for detecting machine defects
Stig Erik Daniel Strömbergsson, Älvsbyn (SE)
Assigned to Aktiebolaget SKF, Gothenburg (SE)
Filed by Aktiebolaget SKF, Gothenburg (SE)
Filed on Jul. 20, 2021, as Appl. No. 17/380,106.
Claims priority of application No. 102020211196.0 (DE), filed on Sep. 7, 2020.
Prior Publication US 2022/0074391 A1, Mar. 10, 2022
Int. Cl. F03D 17/00 (2016.01); F03D 80/50 (2016.01); G01M 13/045 (2019.01); G01M 99/00 (2011.01); G05B 23/00 (2006.01); G06N 3/02 (2006.01); G06N 20/00 (2019.01)
CPC F03D 17/00 (2016.05) [F03D 80/50 (2016.05); G01M 13/045 (2013.01); G01M 99/005 (2013.01); G05B 23/00 (2013.01); G06N 3/02 (2013.01); G06N 20/00 (2019.01); F05B 2260/80 (2013.01); F05B 2270/709 (2013.01)] 6 Claims
OG exemplary drawing
 
1. A method for detecting at least one machine defect, the method comprising:
a) defining from the machine kinematic data at least one condition indicator reflecting the condition of the machine, with respect to a defect to be monitored,
b) recording operating condition data of the machine and condition monitoring data of the machine during a predetermined time period during which the machine is operating normally and determining condition indicator values using the condition monitoring data,
c) training a machine learning algorithm to establish a relation between the operating condition data and the condition indicator values recorded during the predetermined time period,
d) recording current condition monitoring data and determining current condition indicator values from the at least one condition indicator and the current condition monitoring data,
e) predicting condition indicator values with respect to the current operating condition data by the machine learning algorithm,
f) comparing the current condition indicator values and the predicted condition indicator values to determine a percentage of the current condition indicator values which are within a predetermined deviation of the predicted indicator values, and
g) determining the machine's operating status by determining if the percentage of the current condition indicator values which are within the predetermined deviation of the predicted indicator values is less than or equal a predetermined percentage limit which indicates that the machine is presumed to be operating normally.