Early detection of lean blow out (LBO) via generalized D-Markov machine construction

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

doi:10.1109/acc.2014.6859048

Cited by 9 publications

(11 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the estimate of the mutual information is obtained from Eqs. (16) and (17). The computational complexity for Eq.…”

Section: B Estimation Of Mutual Information With Parzen Windowmentioning

confidence: 99%

“…This phenomenon calls for prediction of blowout phenomena and adoption of appropriate measures to mitigate it. Experiments [17] were conducted in a laboratory-scale swirl-stabilized dump combustor that represented a generic gas turbine combustor. Multiple experiments were conducted with liquefied petroleum gas (LPG) fuel at airflow rates of 150, 175 and 200 liters per minute (lpm) for three different fuel-air premixing lengths (i.e., distance of fuel injection port from the dump plane) of L f uel = 35 cm, 25 cm, and 15 cm for Port 1, Port 3, and Port 5, respectively.…”

Section: B Lean Blow-out Prediction In Combustion Systemmentioning

confidence: 99%

See 1 more Smart Citation

Alphabet size selection for symbolization of dynamic data-driven systems: An information-theoretic approach

Sarkar

Chattopdhyay

Ray

et al. 2015

2015 American Control Conference (ACC)

Self Cite

View full text Add to dashboard Cite

Symbolic time series analysis (STSA) is built upon the concept of symbolic dynamics that deals with discretization of dynamical systems in both space and time. The notion of STSA has led to the development of a pattern recognition tool in the paradigm of dynamic data-driven application systems (DDDAS), where a time series of sensor signals is partitioned to obtain a symbol sequence that, in turn, leads to the construction of probabilistic finite state automata (PFSA). Although modeling of PFSA from symbol sequences has been widely reported, similar efforts have not been expended to investigate how to find an appropriate alphabet size for partitioning of time series so that the symbol sequences can be optimally generated. This paper addresses this critical issue and proposes an information-theoretic procedure of data partitioning to extract low-dimensional features from time series. The key idea lies in optimal partitioning of the time series via maximization of the mutual information between the input state probability vector and pattern classes. The proposed procedure has been validated by two examples. The first example elucidates the underlying concept of data partitioning for parameter identification in a Duffing system with a sinusoidal input excitation. The second example is built upon time series of chemiluminescence data to predict lean blow-out (LBO) phenomena in a laboratory-scale combustor. Classification performance of data partitioning is analyzed in each of the two examples.

show abstract

“…Finally, the estimate of the mutual information is obtained from Eqs. (16) and (17). The computational complexity for Eq.…”

Section: B Estimation Of Mutual Information With Parzen Windowmentioning

confidence: 99%

Section: B Lean Blow-out Prediction In Combustion Systemmentioning

confidence: 99%

Alphabet size selection for symbolization of dynamic data-driven systems: An information-theoretic approach

Sarkar

Chattopdhyay

Ray

et al. 2015

2015 American Control Conference (ACC)

Self Cite

View full text Add to dashboard Cite

show abstract

“…The concept of STSA has been used for anomaly detection in physical systems as reported in [110,109,119]. Recently, STSA is applied on pressure and chemiluminescence time series for early detection of Lean-blow out [95,123] and thermo-acoustic instability [108].…”

Section: A B C D E a B C D E A B C D E A B C D E A B C D E A B C D E mentioning

confidence: 99%

“…In this chapter, depth of the D-Markov machine is chosen to be one and it results in the equality of state transition matix (Π) and probability morph matrixπ. Depth greater than one can also be chosen via applying generalized D-Markov machine construction[123,94]. Π is considered as the output feature of the D-Markov machine, which represents the time-series in reduced dimension.…”

mentioning

confidence: 99%