Systematic reduction of chemical mechanisms via rate-controlled constrained equilibrium
File(s)
Author(s)
Koniavitis, Panagiotis
Type
Thesis
Abstract
The development of reduced chemical mechanisms in a systematic way has emerged as a
potential solution to the problem of incorporating the increasingly large chemical
mechanisms into turbulent combustion CFD codes. In this work, Rate-Controlled Constrained
Equilibrium (RCCE) is linked with different systematic methods for identifying the major
species in different detailed mechanisms. As a result, a methodology is proposed for
developing reduced mechanisms with RCCE via a Computational Singular Perturbation (CSP)
analysis of counterflow non-premixed flamelets for different scalar dissipation rates, by
integrating over mixture fraction space a modified CSP pointer and weighting the local
ordering for each strain rate.
RCCE simulations with the derived reduced mechanisms for methane with 16 species and for
propane with 27 species are compared with the integration of the detailed mechanisms GRI
1.2 and USC-Mech-II respectively. The last, and most complex, problem under investigation
is a detailed mechanism for a realistic surrogate fuel for kerosene, for which two reduced
mechanisms are developed via the RCCE-CSP methodology with 17 and 42 species.
The applicability of the methodology is demonstrated in non-premixed flames for
several strain rates, in non-premixed flames ignited with a pilot to
test the dynamic behaviour, in premixed flames for
different equivalence ratios and subsequently in perfectly stirred reactors for
ignition delay times for varying temperature, pressure and equivalence ratio.
An additional combustion regime, studied here, is a laminar non-premixed flamelet,
with time-dependent strain rate at high pressure, to
test the dynamic behaviour of the reduced mechanisms compared to the full model.
Overall very good agreement is obtained, indicating that the methodology can
produce reliable mechanisms for different fuels and for a wide range of
conditions, combining a remarkable degree of reduction in computational time.
potential solution to the problem of incorporating the increasingly large chemical
mechanisms into turbulent combustion CFD codes. In this work, Rate-Controlled Constrained
Equilibrium (RCCE) is linked with different systematic methods for identifying the major
species in different detailed mechanisms. As a result, a methodology is proposed for
developing reduced mechanisms with RCCE via a Computational Singular Perturbation (CSP)
analysis of counterflow non-premixed flamelets for different scalar dissipation rates, by
integrating over mixture fraction space a modified CSP pointer and weighting the local
ordering for each strain rate.
RCCE simulations with the derived reduced mechanisms for methane with 16 species and for
propane with 27 species are compared with the integration of the detailed mechanisms GRI
1.2 and USC-Mech-II respectively. The last, and most complex, problem under investigation
is a detailed mechanism for a realistic surrogate fuel for kerosene, for which two reduced
mechanisms are developed via the RCCE-CSP methodology with 17 and 42 species.
The applicability of the methodology is demonstrated in non-premixed flames for
several strain rates, in non-premixed flames ignited with a pilot to
test the dynamic behaviour, in premixed flames for
different equivalence ratios and subsequently in perfectly stirred reactors for
ignition delay times for varying temperature, pressure and equivalence ratio.
An additional combustion regime, studied here, is a laminar non-premixed flamelet,
with time-dependent strain rate at high pressure, to
test the dynamic behaviour of the reduced mechanisms compared to the full model.
Overall very good agreement is obtained, indicating that the methodology can
produce reliable mechanisms for different fuels and for a wide range of
conditions, combining a remarkable degree of reduction in computational time.
Version
Open Access
Date Issued
2017-09
Date Awarded
2018-03
Copyright Statement
Attribution NoDerivatives 4.0 International Licence (CC BY-ND)
Advisor
Rigopoulos, Stelios
Jones, W. P.
Sponsor
European Union
Grant Number
620143
Publisher Department
Mechanical Engineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)