TR2019-057

Fault Detection and Classification of Time Series Using Localized Matrix Profiles

    •  Zhang, J., Nikovski, D.N., Lee, T.-Y., Fujino, T., "Fault Detection and Classification of Time Series Using Localized Matrix Profiles", IEEE International Conference on Prognostics and Health Management, DOI: 10.1109/​ICPHM.2019.8819389, June 2019.
      BibTeX TR2019-057 PDF
      • @inproceedings{Zhang2019jun,
      • author = {Zhang, Jing and Nikovski, Daniel N. and Lee, Teng-Yok and Fujino, Tomoya},
      • title = {{Fault Detection and Classification of Time Series Using Localized Matrix Profiles}},
      • booktitle = {IEEE International Conference on Prognostics and Health Management},
      • year = 2019,
      • month = jun,
      • doi = {10.1109/ICPHM.2019.8819389},
      • url = {https://www.merl.com/publications/TR2019-057}
      • }
  • MERL Contact:
  • Research Area:

    Data Analytics

Abstract:

We introduce a new primitive, called the Localized Matrix Profile (LMP), for time series data mining. We devise fast algorithms for LMP computation, and propose a fault detector and a fault classifier based on the LMP. A case study using synthetic sensor data generated from a physical model of an electrical motor is provided to demonstrate the effectiveness and efficiency of our approach.