Not Logged In

Detecting the Onset of Machine Failure Using Anomaly Detection Methods

During the lifetime of any machine, components will at some point break down and fail due to wear and tear. In this paper we propose a data-driven approach to anomaly detection for early detection of faults for a condition-based maintenance. For the purpose of this study, a belt-driven single degree of freedom robot arm is designed. The robot arm is conditioned on the torque required to move the arm forward and backward, simulating a door opening and closing operation. Typical failures for this system are identified and simulated. Several semi-supervised algorithms are evaluated and compared in terms of their classification performance. We furthermore compare the needed time to train and test each model and their required memory usage. Our results show that the majority of the tested algorithms can achieve a F1-score of more than 0.9. Successfully detecting failures as they begin to occur promises to address key issues in maintenance like safety and cost effectiveness.

Citation

M. Riazi, O. Zaiane, T. Takeuchi, J. Gunther, M. Lipsett. "Detecting the Onset of Machine Failure Using Anomaly Detection Methods". International Conference on Big Data Analytics and Knowledge Discovery (DAWAK), Linz, Austria, pp 3-12, August 2019.

Keywords: Anomaly detection, Fault detection, Predictive Maintenance, Machinery Diagnostics, Onset of failure, Machine learning
Category: In Conference
Web Links: doi
  Springer

BibTeX

@incollection{Riazi+al:DAWAK19,
  author = {Mohammad Riazi and Osmar R. Zaiane and Tomoharu Takeuchi and
    Johannes Gunther and Micheal Lipsett},
  title = {Detecting the Onset of Machine Failure Using Anomaly Detection
    Methods},
  Pages = {3-12},
  booktitle = {International Conference on Big Data Analytics and Knowledge
    Discovery (DAWAK)},
  year = 2019,
}

Last Updated: September 15, 2020
Submitted by Sabina P

University of Alberta Logo AICML Logo