US 12,105,507 B2
Automated diagnosis of augmented acoustic measurement in industrial environments
Bijan Sayyarodsari, Austin, TX (US); Kadir Liano, Pflugerville, TX (US); and Wei Dai, Austin, TX (US)
Assigned to Rockwell Automation Technologies, Inc., Mayfield Heights, OH (US)
Filed by ROCKWELL AUTOMATION TECHNOLOGIES, INC., Mayfield Heights, OH (US)
Filed on Aug. 31, 2021, as Appl. No. 17/463,159.
Prior Publication US 2023/0061688 A1, Mar. 2, 2023
Int. Cl. G05B 23/02 (2006.01); G05B 19/418 (2006.01)
CPC G05B 23/0281 (2013.01) [G05B 19/4183 (2013.01); G05B 19/4184 (2013.01); G05B 23/0218 (2013.01); G05B 23/0289 (2013.01); G05B 23/024 (2013.01); G05B 2219/37337 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A non-transitory computer-readable medium comprising computer-executable instructions that, when executed, are configured to cause a processor to perform operations comprising:
receiving audio data representative of a plurality of sound waves generated by one or more industrial devices, wherein the audio data is acquired by one or more audio sensors disposed in an industrial environment having the one or more industrial devices;
extracting a plurality of features from the audio data, wherein the plurality of features is representative of a portion of the audio data;
identifying a first subset of the plurality of features based on one or more distances between each of the plurality of features in an information space and each known cluster of a plurality of known clusters in the information space exceeding a respective threshold distance, wherein each known cluster corresponds to a mapping between a respective set of known features and the information space, wherein the first subset of the plurality of features corresponds to an unknown cluster outside of the plurality of known clusters, and wherein each feature of a second subset of the plurality of features corresponds to at least one of the plurality of known clusters, and wherein at least one feature of a third subset of the plurality of features is outside of each of the plurality of known clusters and has an additional distance from each of the plurality of known clusters in the information space that is less than the respective threshold distance;
performing a constrained classification operation on each feature of the first subset of the plurality of features based on a constrained clustering model, wherein the constrained classification operation is configured to analyze each feature of the first subset of the plurality of features separately from the plurality of known features using a batch classification learning mode, and wherein analyzing each feature of the first subset of the plurality of features using the batch classification learning mode comprises designating at least a portion of the first subset of the plurality of features as a new cluster based on each feature of the at least a portion of the first subset of the plurality of features having a threshold percentage of matching features;
updating the constrained clustering model based on the at least a portion of the first subset of the plurality of features and the second subset of the plurality of features, while excluding the third subset of the plurality of features, wherein the updated constrained clustering model is configured to:
add the new cluster to the plurality of known clusters; and
update the mapping to map the at least a portion of the first subset of the plurality of features to the new cluster; and
sending one or more commands to the one or more industrial devices to modify one or more operations of the one or more industrial devices based on the new cluster.