Symbolic time-series analysis for anomaly detection in mechanical systems

Amol Khatkhate, Asok Ray, Eric Keller, Shalabh Gupta, Shin C. Chin

    Research output: Contribution to journalArticle

    30 Citations (Scopus)

    Abstract

    This paper examines the efficacy of a novel method for anomaly detection in mechanical systems, which makes use of a hidden Markov model, derived from the time-series data of pertinent measurement(s). The core concept of the anomaly detection method is symbolic time-series analysis that is built upon the principles of Automata Theory, Information Theory, and Pattern Recognition. The performance of this method is compared with that of other existing pattern-recognition techniques from the perspective of early detection of small fatigue cracks in ductile alloy structures. The experimental apparatus, on which the anomaly detection method is tested, is a multi-degree-of-freedom mass-beam structure excited by oscillatory motion of two electromagnetic shakers. The evolution of fatigue crack damage at one or more failure sites are detected from symbolic time-series analysis of displacement sensor signals.

    Original languageEnglish (US)
    Pages (from-to)439-446
    Number of pages8
    JournalIEEE/ASME Transactions on Mechatronics
    Volume11
    Issue number4
    DOIs
    StatePublished - Dec 1 2006

    Fingerprint

    Time series analysis
    Pattern recognition
    Automata theory
    Information theory
    Hidden Markov models
    Time series
    Sensors
    Fatigue cracks

    All Science Journal Classification (ASJC) codes

    • Control and Systems Engineering
    • Computer Science Applications
    • Electrical and Electronic Engineering

    Cite this

    Khatkhate, Amol ; Ray, Asok ; Keller, Eric ; Gupta, Shalabh ; Chin, Shin C. / Symbolic time-series analysis for anomaly detection in mechanical systems. In: IEEE/ASME Transactions on Mechatronics. 2006 ; Vol. 11, No. 4. pp. 439-446.
    @article{d0280cf061bc4601bd93cfecb4e54cf7,
    title = "Symbolic time-series analysis for anomaly detection in mechanical systems",
    abstract = "This paper examines the efficacy of a novel method for anomaly detection in mechanical systems, which makes use of a hidden Markov model, derived from the time-series data of pertinent measurement(s). The core concept of the anomaly detection method is symbolic time-series analysis that is built upon the principles of Automata Theory, Information Theory, and Pattern Recognition. The performance of this method is compared with that of other existing pattern-recognition techniques from the perspective of early detection of small fatigue cracks in ductile alloy structures. The experimental apparatus, on which the anomaly detection method is tested, is a multi-degree-of-freedom mass-beam structure excited by oscillatory motion of two electromagnetic shakers. The evolution of fatigue crack damage at one or more failure sites are detected from symbolic time-series analysis of displacement sensor signals.",
    author = "Amol Khatkhate and Asok Ray and Eric Keller and Shalabh Gupta and Chin, {Shin C.}",
    year = "2006",
    month = "12",
    day = "1",
    doi = "10.1109/TMECH.2006.878544",
    language = "English (US)",
    volume = "11",
    pages = "439--446",
    journal = "IEEE/ASME Transactions on Mechatronics",
    issn = "1083-4435",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    number = "4",

    }

    Symbolic time-series analysis for anomaly detection in mechanical systems. / Khatkhate, Amol; Ray, Asok; Keller, Eric; Gupta, Shalabh; Chin, Shin C.

    In: IEEE/ASME Transactions on Mechatronics, Vol. 11, No. 4, 01.12.2006, p. 439-446.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Symbolic time-series analysis for anomaly detection in mechanical systems

    AU - Khatkhate, Amol

    AU - Ray, Asok

    AU - Keller, Eric

    AU - Gupta, Shalabh

    AU - Chin, Shin C.

    PY - 2006/12/1

    Y1 - 2006/12/1

    N2 - This paper examines the efficacy of a novel method for anomaly detection in mechanical systems, which makes use of a hidden Markov model, derived from the time-series data of pertinent measurement(s). The core concept of the anomaly detection method is symbolic time-series analysis that is built upon the principles of Automata Theory, Information Theory, and Pattern Recognition. The performance of this method is compared with that of other existing pattern-recognition techniques from the perspective of early detection of small fatigue cracks in ductile alloy structures. The experimental apparatus, on which the anomaly detection method is tested, is a multi-degree-of-freedom mass-beam structure excited by oscillatory motion of two electromagnetic shakers. The evolution of fatigue crack damage at one or more failure sites are detected from symbolic time-series analysis of displacement sensor signals.

    AB - This paper examines the efficacy of a novel method for anomaly detection in mechanical systems, which makes use of a hidden Markov model, derived from the time-series data of pertinent measurement(s). The core concept of the anomaly detection method is symbolic time-series analysis that is built upon the principles of Automata Theory, Information Theory, and Pattern Recognition. The performance of this method is compared with that of other existing pattern-recognition techniques from the perspective of early detection of small fatigue cracks in ductile alloy structures. The experimental apparatus, on which the anomaly detection method is tested, is a multi-degree-of-freedom mass-beam structure excited by oscillatory motion of two electromagnetic shakers. The evolution of fatigue crack damage at one or more failure sites are detected from symbolic time-series analysis of displacement sensor signals.

    UR - http://www.scopus.com/inward/record.url?scp=33847743017&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=33847743017&partnerID=8YFLogxK

    U2 - 10.1109/TMECH.2006.878544

    DO - 10.1109/TMECH.2006.878544

    M3 - Article

    VL - 11

    SP - 439

    EP - 446

    JO - IEEE/ASME Transactions on Mechatronics

    JF - IEEE/ASME Transactions on Mechatronics

    SN - 1083-4435

    IS - 4

    ER -