Open AccessAutoencoder-based anomaly detector for gear tooth bending fatigue cracks
Author(s) -
Nenad G. Nenadic,
Adrian A. Hood,
Christopher Valant,
Josiah Martuscello,
Patrick Horney,
Allen Jones,
Jared Lantner
Publication year - 2021
Publication title -
proceedings of the annual conference of the prognostics and health management society
Language(s) - English
Resource type - Journals
ISSN - 2325-0178
DOI - 10.36001/phmconf.2021.v13i1.3003
Subject(s) - autoencoder , anomaly (physics) , anomaly detection , baseline (sea) , data set , detector , vibration , set (abstract data type) , bending , root (linguistics) , structural engineering , gear tooth , computer science , artificial intelligence , pattern recognition (psychology) , engineering , acoustics , geology , physics , artificial neural network , telecommunications , oceanography , philosophy , programming language , condensed matter physics , linguistics
The article reports on anomaly detection performance of data-driven models based on a few selected autoencoder topologies and compares them to the performance of a set of popular classical vibration-based condition indicators. The evaluation of these models employed data that consisted of baseline gearbox runs and the associated runs with seeded bending cracks in the root of the gear teeth for eight different gear pairings. The analyses showed that the data-driven models, trained on a subset of baseline data outperformed classical CIs as anomaly detectors.