Lean blowout (LBO) prediction through symbolic time series analysis

Swarnendu Sen, Soumalya Sarkar, Rajendra R. Chaudhari, Achintya Mukhopadhyay, Asok Ray

    Research output: Chapter in Book/Report/Conference proceedingChapter

    2 Citations (Scopus)

    Abstract

    Lean combustion is preferred in combustors for cleaner emission. But, the major problem in running the combustor in lean condition is the flame blow-out, due to lower burning velocity in lean condition. To sustain lean combustion, early prediction of lean blowout (LBO) is needed. This need calls for developing new strategies to predict LBO. This work is on development of strategies based on symbolic analysis of time series data obtained from the combustor. Optical signal (CH* chemiluminescence) based time series data has been captured from a laboratory scale dump combustor. Gaseous fuel has been used at different premixing regimes to run the combustor. The time series data obtained are partitioned to different cells to get symbol strings and subsequently generate a probabilistic finite state automata (PFSA). The constructed PFSA is called D-Markov machines. These machines have a deterministic algebraic structure and the states are represented by symbol blocks of length D or less, where D is a positive integer. The estimated state probability vector is found out on the basis of number of occurrence of each symbol over a time duration. A reference state vector is calculated at a condition that is sufficiently away from the state of LBO. For a given state, the deviation of the state vector from the reference vector can be computed and used as an anomaly measure for early detection of LBO. The problem of LBO prediction is posed in a pattern classification setting and the underlying algorithms have been tested on experimental data at different extents of fuel-air premixing and fuel/air ratio. It is found that, over a wide range of fuel-air premixing, performance of D-Markov machines, as LBO predictors, is better with D > 1 than those with D = 1.

    Original languageEnglish (US)
    Title of host publicationCombustion for Power Generation and Transportation
    Subtitle of host publicationTechnology, Challenges and Prospects
    PublisherSpringer Singapore
    Pages153-167
    Number of pages15
    ISBN (Electronic)9789811037856
    ISBN (Print)9789811037849
    DOIs
    StatePublished - Jan 20 2017

    Fingerprint

    Time series analysis
    Combustors
    Time series
    Finite automata
    Air
    Chemiluminescence
    Pattern recognition

    All Science Journal Classification (ASJC) codes

    • Engineering(all)

    Cite this

    Sen, S., Sarkar, S., Chaudhari, R. R., Mukhopadhyay, A., & Ray, A. (2017). Lean blowout (LBO) prediction through symbolic time series analysis. In Combustion for Power Generation and Transportation: Technology, Challenges and Prospects (pp. 153-167). Springer Singapore. https://doi.org/10.1007/978-981-10-3785-6_8
    Sen, Swarnendu ; Sarkar, Soumalya ; Chaudhari, Rajendra R. ; Mukhopadhyay, Achintya ; Ray, Asok. / Lean blowout (LBO) prediction through symbolic time series analysis. Combustion for Power Generation and Transportation: Technology, Challenges and Prospects. Springer Singapore, 2017. pp. 153-167
    @inbook{a3e32c6a663f41b9814acb60a6894f55,
    title = "Lean blowout (LBO) prediction through symbolic time series analysis",
    abstract = "Lean combustion is preferred in combustors for cleaner emission. But, the major problem in running the combustor in lean condition is the flame blow-out, due to lower burning velocity in lean condition. To sustain lean combustion, early prediction of lean blowout (LBO) is needed. This need calls for developing new strategies to predict LBO. This work is on development of strategies based on symbolic analysis of time series data obtained from the combustor. Optical signal (CH* chemiluminescence) based time series data has been captured from a laboratory scale dump combustor. Gaseous fuel has been used at different premixing regimes to run the combustor. The time series data obtained are partitioned to different cells to get symbol strings and subsequently generate a probabilistic finite state automata (PFSA). The constructed PFSA is called D-Markov machines. These machines have a deterministic algebraic structure and the states are represented by symbol blocks of length D or less, where D is a positive integer. The estimated state probability vector is found out on the basis of number of occurrence of each symbol over a time duration. A reference state vector is calculated at a condition that is sufficiently away from the state of LBO. For a given state, the deviation of the state vector from the reference vector can be computed and used as an anomaly measure for early detection of LBO. The problem of LBO prediction is posed in a pattern classification setting and the underlying algorithms have been tested on experimental data at different extents of fuel-air premixing and fuel/air ratio. It is found that, over a wide range of fuel-air premixing, performance of D-Markov machines, as LBO predictors, is better with D > 1 than those with D = 1.",
    author = "Swarnendu Sen and Soumalya Sarkar and Chaudhari, {Rajendra R.} and Achintya Mukhopadhyay and Asok Ray",
    year = "2017",
    month = "1",
    day = "20",
    doi = "10.1007/978-981-10-3785-6_8",
    language = "English (US)",
    isbn = "9789811037849",
    pages = "153--167",
    booktitle = "Combustion for Power Generation and Transportation",
    publisher = "Springer Singapore",

    }

    Sen, S, Sarkar, S, Chaudhari, RR, Mukhopadhyay, A & Ray, A 2017, Lean blowout (LBO) prediction through symbolic time series analysis. in Combustion for Power Generation and Transportation: Technology, Challenges and Prospects. Springer Singapore, pp. 153-167. https://doi.org/10.1007/978-981-10-3785-6_8

    Lean blowout (LBO) prediction through symbolic time series analysis. / Sen, Swarnendu; Sarkar, Soumalya; Chaudhari, Rajendra R.; Mukhopadhyay, Achintya; Ray, Asok.

    Combustion for Power Generation and Transportation: Technology, Challenges and Prospects. Springer Singapore, 2017. p. 153-167.

    Research output: Chapter in Book/Report/Conference proceedingChapter

    TY - CHAP

    T1 - Lean blowout (LBO) prediction through symbolic time series analysis

    AU - Sen, Swarnendu

    AU - Sarkar, Soumalya

    AU - Chaudhari, Rajendra R.

    AU - Mukhopadhyay, Achintya

    AU - Ray, Asok

    PY - 2017/1/20

    Y1 - 2017/1/20

    N2 - Lean combustion is preferred in combustors for cleaner emission. But, the major problem in running the combustor in lean condition is the flame blow-out, due to lower burning velocity in lean condition. To sustain lean combustion, early prediction of lean blowout (LBO) is needed. This need calls for developing new strategies to predict LBO. This work is on development of strategies based on symbolic analysis of time series data obtained from the combustor. Optical signal (CH* chemiluminescence) based time series data has been captured from a laboratory scale dump combustor. Gaseous fuel has been used at different premixing regimes to run the combustor. The time series data obtained are partitioned to different cells to get symbol strings and subsequently generate a probabilistic finite state automata (PFSA). The constructed PFSA is called D-Markov machines. These machines have a deterministic algebraic structure and the states are represented by symbol blocks of length D or less, where D is a positive integer. The estimated state probability vector is found out on the basis of number of occurrence of each symbol over a time duration. A reference state vector is calculated at a condition that is sufficiently away from the state of LBO. For a given state, the deviation of the state vector from the reference vector can be computed and used as an anomaly measure for early detection of LBO. The problem of LBO prediction is posed in a pattern classification setting and the underlying algorithms have been tested on experimental data at different extents of fuel-air premixing and fuel/air ratio. It is found that, over a wide range of fuel-air premixing, performance of D-Markov machines, as LBO predictors, is better with D > 1 than those with D = 1.

    AB - Lean combustion is preferred in combustors for cleaner emission. But, the major problem in running the combustor in lean condition is the flame blow-out, due to lower burning velocity in lean condition. To sustain lean combustion, early prediction of lean blowout (LBO) is needed. This need calls for developing new strategies to predict LBO. This work is on development of strategies based on symbolic analysis of time series data obtained from the combustor. Optical signal (CH* chemiluminescence) based time series data has been captured from a laboratory scale dump combustor. Gaseous fuel has been used at different premixing regimes to run the combustor. The time series data obtained are partitioned to different cells to get symbol strings and subsequently generate a probabilistic finite state automata (PFSA). The constructed PFSA is called D-Markov machines. These machines have a deterministic algebraic structure and the states are represented by symbol blocks of length D or less, where D is a positive integer. The estimated state probability vector is found out on the basis of number of occurrence of each symbol over a time duration. A reference state vector is calculated at a condition that is sufficiently away from the state of LBO. For a given state, the deviation of the state vector from the reference vector can be computed and used as an anomaly measure for early detection of LBO. The problem of LBO prediction is posed in a pattern classification setting and the underlying algorithms have been tested on experimental data at different extents of fuel-air premixing and fuel/air ratio. It is found that, over a wide range of fuel-air premixing, performance of D-Markov machines, as LBO predictors, is better with D > 1 than those with D = 1.

    UR - http://www.scopus.com/inward/record.url?scp=85019389268&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85019389268&partnerID=8YFLogxK

    U2 - 10.1007/978-981-10-3785-6_8

    DO - 10.1007/978-981-10-3785-6_8

    M3 - Chapter

    SN - 9789811037849

    SP - 153

    EP - 167

    BT - Combustion for Power Generation and Transportation

    PB - Springer Singapore

    ER -

    Sen S, Sarkar S, Chaudhari RR, Mukhopadhyay A, Ray A. Lean blowout (LBO) prediction through symbolic time series analysis. In Combustion for Power Generation and Transportation: Technology, Challenges and Prospects. Springer Singapore. 2017. p. 153-167 https://doi.org/10.1007/978-981-10-3785-6_8