US 11,734,604 B2
Anomaly detection based on device vibration
Alexandru Ardel, Austin, TX (US); Shashank Bassi, Austin, TX (US); Elmira M Bonab, Austin, TX (US); and Jeff Brown, Cedar Park, TX (US)
Assigned to SPARKCOGNITION, INC., Austin, TX (US)
Filed by SparkCognition, Inc., Austin, TX (US)
Filed on Apr. 15, 2020, as Appl. No. 16/849,060.
Prior Publication US 2021/0326741 A1, Oct. 21, 2021
Int. Cl. G06N 20/00 (2019.01); G05D 7/06 (2006.01); G06N 7/00 (2006.01); G06N 7/01 (2023.01)
CPC G06N 20/00 (2019.01) [G05D 7/0676 (2013.01); G06N 7/01 (2023.01)] 39 Claims
OG exemplary drawing
 
1. A method of detecting deviation from an operational state of a rotational device, the method comprising:
receiving, from one or more sensor devices coupled to the rotational device, frequency domain data indicative of vibration data sensed during a sensing period;
processing the frequency domain data using a trained anomaly detection model to generate an anomaly score for the sensing period, wherein processing the frequency domain data comprises:
inputting the frequency domain data to an autoencoder; and
generating a residual based on an output of the autoencoder and the frequency domain data, wherein the anomaly score is based on the residual; and
processing the anomaly score using an alert generation model to determine whether to generate an alert.