| CPC G01M 17/10 (2013.01) [B61K 9/04 (2013.01); B61K 9/12 (2013.01); B61L 25/021 (2013.01); F16C 19/52 (2013.01); F16C 19/527 (2013.01); G01M 13/045 (2013.01); G06N 20/00 (2019.01); F16C 2233/00 (2013.01); F16C 2326/10 (2013.01); H04W 84/18 (2013.01)] | 33 Claims |

|
1. A method for operating a wireless sensor node deployed on a railway asset, comprising:
receiving, by a processor, vibration sensor data from one or more vibration sensors and bearing temperature data from a temperature sensor;
performing, by the processor, wheelset health check operations comprising:
using the vibration sensor data as an input to a machine learning model that is trained to detect vibration patterns predictive of future failures of both a bearing portion of the wheelset and a wheel portion of the wheelset; and
in the case that none of the vibration patterns were detected by the machine learning model, using the bearing temperature data as another input to the machine learning model that is trained to detect a fault or defect in the bearing portion of the wheelset; and
altering, by the processor, a behavior of the wireless sensor node based on a result from performance of the wheelset health check operations.
|