1119 Chemiluminescence intensity variations of a combustion instability in a lean premixed gas-turbine model combustor

2013 ◽  
Vol 2013.88 (0) ◽  
pp. _11-29_
Author(s):  
Taku KURIYAMA ◽  
Yuta SHINODA ◽  
Hiroshi GOTODA ◽  
Shigeru TACHIBANA
2012 ◽  
Vol 22 (4) ◽  
pp. 043128 ◽  
Author(s):  
Hiroshi Gotoda ◽  
Masahito Amano ◽  
Takaya Miyano ◽  
Takuya Ikawa ◽  
Koshiro Maki ◽  
...  

Author(s):  
Hiroshi Gotoda ◽  
Kenta Hayashi ◽  
Ryosuke Tsujimoto ◽  
Shohei Domen ◽  
Shigeru Tachibana

We present an experimental study on the nonlinear dynamics of combustion instability in a lean premixed gas-turbine model combustor with a swirl-stabilized turbulent flame. Intermittent combustion oscillations switching irregularly back and forth between burst and pseudo-periodic oscillations exhibit the deterministic nature of chaos. This is clearly demonstrated by considering two nonlinear forecasting methods: an extended version (Gotoda et al., 2015, “Nonlinear Forecasting of the Generalized Kuramoto-Sivashinsky Equation,” Int. J. Bifurcation Chaos, 25, p. 1530015) of the Sugihara and May algorithm (Sugihara and May, 1990, “Nonlinear Forecasting as a Way of Distinguishing Chaos From Measurement Error in Time Series,” Nature, 344, pp. 734–741) as a local predictor, and a generalized radial basis function (GRBF) network as a global predictor (Gotoda et al., 2012, “Characterization of Complexities in Combustion Instability in a Lean Premixed Gas-Turbine Model Combustor,” Chaos, 22, p. 043128; Gotoda et al., 2016 (unpublished)). The former enables us to extract the short-term predictability and long-term unpredictability of chaos, while the latter can produce surrogate data to test for determinism by a free-running approach. The permutation entropy based on a symbolic sequence approach is estimated for the surrogate data to test for determinism and is also used as an online detector to prevent lean blowout.


Author(s):  
L. Rosentsvit ◽  
Y. Levy ◽  
V. Erenburg ◽  
V. Sherbaum ◽  
V. Ovcharenko ◽  
...  

The present work is concerned with improving combustion stability in lean premixed (LP) gas turbine combustors by injecting free radicals into the combustion zone. The work is a joint experimental and numerical effort aimed at investigating the feasibility of incorporating a circumferential pilot combustor, which operates under rich conditions and directs its radicals enriched exhaust gases into the main combustion zone as the means for stabilization. The investigation includes the development of a chemical reactors network (CRN) model that is based on perfectly stirred reactors modules and on preliminary CFD analysis as well as on testing the method on an experimental model under laboratory conditions. The study is based on the hypothesis that under lean combustion conditions, combustion instability is linked to local extinctions of the flame and consequently, there is a direct correlation between the limiting conditions affecting combustion instability and the lean blowout (LBO) limit of the flame. The experimental results demonstrated the potential reduction of the combustion chamber's LBO limit while maintaining overall NOx emission concentration values within the typical range of low NOx burners and its delicate dependence on the equivalence ratio of the ring pilot flame. A similar result was revealed through the developed CHEMKIN-PRO CRN model that was applied to find the LBO limits of the combined pilot burner and main combustor system, while monitoring the associated emissions. Hence, both the CRN model, and the experimental results, indicate that the radicals enriched ring jet is effective at stabilizing the LP flame, while keeping the NOx emission level within the characteristic range of low NOx combustors.


2010 ◽  
Vol 52 (3) ◽  
pp. 555-567 ◽  
Author(s):  
Isaac Boxx ◽  
Christoph M. Arndt ◽  
Campbell D. Carter ◽  
Wolfgang Meier

2011 ◽  
Vol 21 (1) ◽  
pp. 013124 ◽  
Author(s):  
Hiroshi Gotoda ◽  
Hiroyuki Nikimoto ◽  
Takaya Miyano ◽  
Shigeru Tachibana

Sign in / Sign up

Export Citation Format

Share Document