Early Detection of Thermoacoustic Instabilities Using Hidden Markov Models

Sudeepta Mondal, Najah F. Ghalyan, Asok Ray, Achintya Mukhopadhyay

    Research output: Contribution to journalArticle

    Abstract

    This paper presents a dynamic data-driven method for early detection of thermoacoustic instabilities in combustors based on short-length time series of sensor data, where the objective is near-real-time monitoring and active control of pressure oscillations. The main idea is to use the available data at different regimes of the combustion process to train respective hidden-variable models using the concept of Hidden Markov Modeling (HMM) as a statistical learning tool; here, (short-length) time-series data of pressure oscillations are used to infer a Markov chain with unobserved (hidden) states. The proposed HMM-based method has been validated on experimental data collected from an electrically heated Rijke tube apparatus for predicting onset of thermoacoustic instabilities. The results have been compared with those of the current state-of-the-art measurement techniques for instability growth rate and associated computational complexity. The applicability of the proposed method has been demonstrated with respect to anomaly detection and regime identification with limited data requirements, making it a potential candidate for monitoring and active control of thermoacoustic instabilities in commercial-scale combustors.

    Original languageEnglish (US)
    Pages (from-to)1309-1336
    Number of pages28
    JournalCombustion science and technology
    Volume191
    Issue number8
    DOIs
    StatePublished - Aug 3 2019

    Fingerprint

    Thermoacoustics
    Hidden Markov models
    pressure oscillations
    active control
    combustion chambers
    Combustors
    Time series
    Monitoring
    Markov chains
    Markov processes
    learning
    Computational complexity
    anomalies
    tubes
    requirements
    sensors
    Sensors

    All Science Journal Classification (ASJC) codes

    • Chemistry(all)
    • Chemical Engineering(all)
    • Fuel Technology
    • Energy Engineering and Power Technology
    • Physics and Astronomy(all)

    Cite this

    Mondal, Sudeepta ; Ghalyan, Najah F. ; Ray, Asok ; Mukhopadhyay, Achintya. / Early Detection of Thermoacoustic Instabilities Using Hidden Markov Models. In: Combustion science and technology. 2019 ; Vol. 191, No. 8. pp. 1309-1336.
    @article{9d8eab35e2b34412b4bf6e253a3f6b22,
    title = "Early Detection of Thermoacoustic Instabilities Using Hidden Markov Models",
    abstract = "This paper presents a dynamic data-driven method for early detection of thermoacoustic instabilities in combustors based on short-length time series of sensor data, where the objective is near-real-time monitoring and active control of pressure oscillations. The main idea is to use the available data at different regimes of the combustion process to train respective hidden-variable models using the concept of Hidden Markov Modeling (HMM) as a statistical learning tool; here, (short-length) time-series data of pressure oscillations are used to infer a Markov chain with unobserved (hidden) states. The proposed HMM-based method has been validated on experimental data collected from an electrically heated Rijke tube apparatus for predicting onset of thermoacoustic instabilities. The results have been compared with those of the current state-of-the-art measurement techniques for instability growth rate and associated computational complexity. The applicability of the proposed method has been demonstrated with respect to anomaly detection and regime identification with limited data requirements, making it a potential candidate for monitoring and active control of thermoacoustic instabilities in commercial-scale combustors.",
    author = "Sudeepta Mondal and Ghalyan, {Najah F.} and Asok Ray and Achintya Mukhopadhyay",
    year = "2019",
    month = "8",
    day = "3",
    doi = "10.1080/00102202.2018.1523900",
    language = "English (US)",
    volume = "191",
    pages = "1309--1336",
    journal = "Combustion Science and Technology",
    issn = "0010-2202",
    publisher = "Taylor and Francis Ltd.",
    number = "8",

    }

    Early Detection of Thermoacoustic Instabilities Using Hidden Markov Models. / Mondal, Sudeepta; Ghalyan, Najah F.; Ray, Asok; Mukhopadhyay, Achintya.

    In: Combustion science and technology, Vol. 191, No. 8, 03.08.2019, p. 1309-1336.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Early Detection of Thermoacoustic Instabilities Using Hidden Markov Models

    AU - Mondal, Sudeepta

    AU - Ghalyan, Najah F.

    AU - Ray, Asok

    AU - Mukhopadhyay, Achintya

    PY - 2019/8/3

    Y1 - 2019/8/3

    N2 - This paper presents a dynamic data-driven method for early detection of thermoacoustic instabilities in combustors based on short-length time series of sensor data, where the objective is near-real-time monitoring and active control of pressure oscillations. The main idea is to use the available data at different regimes of the combustion process to train respective hidden-variable models using the concept of Hidden Markov Modeling (HMM) as a statistical learning tool; here, (short-length) time-series data of pressure oscillations are used to infer a Markov chain with unobserved (hidden) states. The proposed HMM-based method has been validated on experimental data collected from an electrically heated Rijke tube apparatus for predicting onset of thermoacoustic instabilities. The results have been compared with those of the current state-of-the-art measurement techniques for instability growth rate and associated computational complexity. The applicability of the proposed method has been demonstrated with respect to anomaly detection and regime identification with limited data requirements, making it a potential candidate for monitoring and active control of thermoacoustic instabilities in commercial-scale combustors.

    AB - This paper presents a dynamic data-driven method for early detection of thermoacoustic instabilities in combustors based on short-length time series of sensor data, where the objective is near-real-time monitoring and active control of pressure oscillations. The main idea is to use the available data at different regimes of the combustion process to train respective hidden-variable models using the concept of Hidden Markov Modeling (HMM) as a statistical learning tool; here, (short-length) time-series data of pressure oscillations are used to infer a Markov chain with unobserved (hidden) states. The proposed HMM-based method has been validated on experimental data collected from an electrically heated Rijke tube apparatus for predicting onset of thermoacoustic instabilities. The results have been compared with those of the current state-of-the-art measurement techniques for instability growth rate and associated computational complexity. The applicability of the proposed method has been demonstrated with respect to anomaly detection and regime identification with limited data requirements, making it a potential candidate for monitoring and active control of thermoacoustic instabilities in commercial-scale combustors.

    UR - http://www.scopus.com/inward/record.url?scp=85054505378&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85054505378&partnerID=8YFLogxK

    U2 - 10.1080/00102202.2018.1523900

    DO - 10.1080/00102202.2018.1523900

    M3 - Article

    AN - SCOPUS:85054505378

    VL - 191

    SP - 1309

    EP - 1336

    JO - Combustion Science and Technology

    JF - Combustion Science and Technology

    SN - 0010-2202

    IS - 8

    ER -