Chaos in an imperfectly premixed model combustor

Kabiraj, Lipika; Saurabh, Aditya; Karimi, Nader; Sailor, Anna; Mastorakos, Epaminondas; Dowling, Ann P.; Paschereit, Christian Oliver

doi:10.1063/1.4906943

Cited by 63 publications

(23 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The increased proportion of short line segments-deterministic structures-interrupted by isolated points or white horizontal and vertical spaces is clearly visible. Short line segments indicate a positive maximal Lyapunov exponent 18,19 , and hence, chaotic dynamics of the thermoacoustic coupling, that have also been identified in other studies 15,28,30 . In order to describe the distribution of line segments in the RP, a range of line lengths will be required.…”

Section: Reacting Flow: Recurrence Plotsmentioning

confidence: 82%

Characterizing the signature of flame flashback precursor through recurrence analysis

Christodoulou

Kabiraj

Saurabh

et al. 2016

Chaos: An Interdisciplinary Journal of Nonlinear Science

Self Cite

View full text Add to dashboard Cite

In this paper it is shown that prior to flashback, small dynamical changes appear in the system. These changes appear as a drift in the recurrence plots and are found to be associated with a gradual increase in the determinism and recurrence rate. Thus, this study indicates that precursors to flame flashback exist and can be detected in the multidimensional phase space reconstructed from pressure measurements acquired during flashback. This observation could have broad academic as well as industrial implications.PACS numbers: 05.45. Tp, 47.70.Pq Flame flashback is a complex phenomenon resulting from interactions involving multiple physical processes, each governed by its own inherent temporal and spatial scale. In addition, flashback is also a transient event, with the actual event lasting only fractions of a second. Investigation of flashback requires multiple, simultaneous diagnostics without prior knowledge of the relevant time and length scales of the physical processes involved. Detection, accordingly, deals with postevent characterization. In this paper, an attempt to detect subtle dynamics prior to flashbackwhich linear, time and frequency-domain methods cannot reveal-is made through recurrence analysis of the multidimensional phase space reconstruction from pressure time traces. Recurrence analysis is superior to the common techniques (e.g. spectra) used for the analysis of short, noisy datasets for the purpose of identifying dynamical changes. Following conventional recurrence analysis methods, different sections of pressure time traces from reacting flow conditions are analyzed. Recurrence plots and recurrence quantification measures immediately before flashback are compared to those long before flashback during the normal operation of the combustor. A comparison between recurrence characteristics at isothermal flow conditions and at reacting flow conditions is included to a) highlight the dynamics of the thermoacoustically unstable condition at which flashback occurs and b) to establish the limits of fluctuations in recurrence quantification analysis measures. It is observed that small drifts exist in the dynamical behavior of the system before flashback and are made apparent by recurrence methods. It is found that the changes span 2-4 seconds prior to flashback. Within this time a) Electronic address: l.christodoulou.1@research.gla.ac.uk b) Electronic address: lipika.kabiraj@tu-berlin.de frame, determinism and recurrence rate increase gradually as flashback is approached.

show abstract

Section: Reacting Flow: Recurrence Plotsmentioning

confidence: 82%

Characterizing the signature of flame flashback precursor through recurrence analysis

Christodoulou

Kabiraj

Saurabh

et al. 2016

Chaos: An Interdisciplinary Journal of Nonlinear Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…In thermoacoustic combustors, random noise n 3 (t) due to entropy fluctuations 36,37 unsteady combustion such as combustion-driven oscillations [38][39][40] or is present. Thus the noisy thermoacoustic model is described by…”

Section: The Nonlinear Thermoacoustic Modelmentioning

confidence: 99%

Effects of background noises on nonlinear dynamics of a modelled thermoacoustic combustor

Zhao

2018

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

In this work, the effects of background noises on nonlinear dynamics of a modelled standing-wave thermoacoustic system with subcritical Hopf bifurcation behaviors are studied. These noises include (1) pressure-coupled (acoustic), (2) velocity-coupled (flow), and (3) external combustion noise. It is found that these three types of background noises play important, but different roles in changing the hysteresis width and stability boundary. In addition, the stochastic transition from stability to instability is investigated, as the noise intensity is varied. Two different stochastic P-bifurcations are identified. One is related to a craterlike probability density distribution. The other is associated with a probability density distribution characterized with two peaks and one trough. With each type of noise affecting the system's stochastic behaviors being evaluated, the effect of two different types of noises is then studied. It is shown that the combined noises (types 1 and 2) cannot only destabilize global stable thermoacoustic system, but also stabilize linearly unstable system. This depends strongly on the superimposition form of the two types of noises. In addition, when the thermoacoustic system is disturbed by the combined noise (types 3 and 1 or types 3 and 2), the transition process is dominated by the combustion noise.

show abstract

“…It is lag-reconstructed from a time series p. The embedding dimension M is determined by the false nearest neighbor algorithm [35]. And the time delay parameter τ is estimated with mutual information method [34]. The un-threshold recurrence (UR) plot represents a symmetric distance matrix between the state vectors, i.e.…”

Section: Recurrence Plot Of the Measured Pressure Signalmentioning

confidence: 99%

“…p(i) and p(j) are vectors in the state space and Θ is the Heaviside function. Recurrence quantification analysis (RQA) [35] can not only characterize small-structures (e.g., small dots, vertical and diagonal lines) in the recurrence plots but also quantify the system chaos-order or chaos-chaos transitions [34,35]. The recurrence quantifiers include (1) determinism, (2) recurrence rate, entropy, and trapping time.…”

Section: Recurrence Plot Of the Measured Pressure Signalmentioning

confidence: 99%

“…The scientific study on turbulent combustion noise followed the previous researches on turbulent flows and flames [24,25]. Nonlinear means based on the phase space characterization [26,27] are increasingly being combined in investigations on laminar [28] and turbulent combustion [29][30][31][32][33][34], internal combustion engines [35] as well as related fields [36,37]. These techniques allow studies through an alternate perspective: the phase space.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Experimental study of equivalence ratio and fuel flow rate effects on nonlinear thermoacoustic instability in a swirl combustor

Zhao

et al. 2017

Applied Energy

View full text Add to dashboard Cite

Industrial combustion systems such as power generation gas turbines, rocket motors, furnaces and boilers often face the problem of large-amplitude self-excited pressure oscillations that occur due to the onset of thermoacoustic instability. To prevent the onset of such instability, understanding the effects of fuel-air equivalence ratio ϕ, fuel flow rate V f on self-excited nonlinear thermoacoustic oscillations is of fundamental and practical importance. Experimental investigation of the roles of these parameters on triggering thermoacoustic instability in a swirl combustor has received very little attention. In this work, we design a swirling thermocoustic combustor and conduct a series of experimental tests. Autocorrelation and recurrence analysis of phase space trajectories reconstructed from the acoustic pressure time trace are performed. These experimental tests allow us to study the effect of fuel-air equivalence ratio ϕ on the onset of thermoacoustic instability by varying the fuel volume flow rate V f . We demonstrate that the fuel volume flow rate and the equivalence ratio play different but critical roles on generating thermoacoustic instability at different frequencies and amplitudes. Maximum sound pressure level can be as high as 135 dB. In addition, mode switching, (i.e. frequency swap) is found to occur between approximately ω 3 ≈ 510 Hz and ω 1 ≈ 170 Hz, depending on the equivalence ratio ϕ. Furthermore, the dominant frequency corresponding to the maximum amplitude is shown to be shifted by approximately 20%, as the fuel flow rate V f is increased and the combustion condition is changed from lean to rich. These findings are quite useful for designing a feedback control strategy to stabilize an unstable combustor. The present work opens up an applicable means to design a stable swirling combustor.

show abstract

Chaos in an imperfectly premixed model combustor

Cited by 63 publications

References 33 publications

Characterizing the signature of flame flashback precursor through recurrence analysis

Characterizing the signature of flame flashback precursor through recurrence analysis

Effects of background noises on nonlinear dynamics of a modelled thermoacoustic combustor

Experimental study of equivalence ratio and fuel flow rate effects on nonlinear thermoacoustic instability in a swirl combustor

Contact Info

Product

Resources

About