Tandem Connectionist Anomaly Detection: Use of Faulty Vibration Signals in Feature Representation Learning

Takanori Hasegawa, Jun Ogata, Masahiro Murakawa, Tetsuji Ogawa

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    1 Citation (Scopus)

    Abstract

    An effective use of faulty-state data is proposed to achieve robust, accurate data-driven anomaly (fault) detection for rotating machine. Although using faulty data in the training process generally can improve the performance of anomaly detection system, it is rare to obtain enough samples to train failures or defects on a target machine. We therefore utilize the existing data from non-target (different-type) machines for feature representation learning to improve anomaly detection for the target machine. Specifically, deep neural networks (DNNs) that are trained to discriminate the normal and faulty states of the non-target machines are used to extract features. The extracted features are then taken as inputs to an anomaly detector based on Gaussian mixture models (GMMs). This architecture is called DNN/GMM tandem connectionist anomaly detection. Experimental comparisons using vibration signals from actual wind turbine components demonstrated that the developed tandem connectionist system yielded significant improvements over existing systems, and that the representation learning performed robustly with respect to differences in machine types.

    Original languageEnglish
    Title of host publication2018 IEEE International Conference on Prognostics and Health Management, ICPHM 2018
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    ISBN (Electronic)9781538611647
    DOIs
    Publication statusPublished - 2018 Aug 27
    Event2018 IEEE International Conference on Prognostics and Health Management, ICPHM 2018 - Seattle, United States
    Duration: 2018 Jun 112018 Jun 13

    Publication series

    Name2018 IEEE International Conference on Prognostics and Health Management, ICPHM 2018

    Conference

    Conference2018 IEEE International Conference on Prognostics and Health Management, ICPHM 2018
    CountryUnited States
    CitySeattle
    Period18/6/1118/6/13

      Fingerprint

    Keywords

    • Anomaly detection
    • Condition monitoring
    • Data-driven method
    • Machine learning
    • Representation learning
    • Vibration signals
    • Wind turbine

    ASJC Scopus subject areas

    • Statistics, Probability and Uncertainty
    • Civil and Structural Engineering
    • Safety, Risk, Reliability and Quality

    Cite this

    Hasegawa, T., Ogata, J., Murakawa, M., & Ogawa, T. (2018). Tandem Connectionist Anomaly Detection: Use of Faulty Vibration Signals in Feature Representation Learning. In 2018 IEEE International Conference on Prognostics and Health Management, ICPHM 2018 [8448450] (2018 IEEE International Conference on Prognostics and Health Management, ICPHM 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICPHM.2018.8448450