Diagnosing Local Faults in Gearbox & Bearing Systems for Condition-Based Maintenance (CBM) Applications

A. Nadkarni, J. Hofmeister, W. Pena, C. Curti
Ridgetop Group Inc., Arizona, United States

Poster stand number: W146

Keywords: IoT, Vibration Analysis, Condition-Based Maintenance (CBM), Prognostics, PHM

Gears and bearings, key components of every power transmission system, are often subjected to excessive faults like impacts and cracks that result in excessive vibration, noise, unwanted energy loss, and eventually failure. In the last three decades, a lot of effort has been devoted toward diagnosing such faults at early stages using vibration signals from an in-operation rotating system, to prevent catastrophic breakdowns. This ongoing experiment is a proof-of-concept for detecting, localizing, and assessing any local faults in rotating system using our in-house testbed. Vibration signals are collected using ultra-low powered IoT RotoSense™ sensors mounted to commercial-off-the-shelf accelerometers on the bearing and shaft of the rotating testbed. Followingly, the extracted condition-based feature data are processed using ARULE™ - Adaptive Remaining Useful Life Estimator, RGI’s robust prognostic reasoner, to estimate key prognostic quantities like the Remaining Useful Life (RUL), State of Health (SoH), and the Prognostic Horizon (PH) that delineate the system’s condition, life, and maintenance status.