A bluff-body combustor, with recirculation zone and simple boundary conditions, is ideal as a compromise for an industrial combustor for validating combustion models. This combustor, however, has proved to be very challenging to the combustion modellers in a number of previous studies. In the present study, an improved prediction has been reported through better representation of turbulence effect by Reynolds stress transport model and extended upstream computational domain. Thermo-chemical properties of the flame have been represented by a laminar flamelet model. A comparison among reduced chemical kinetic mechanism of Peters and detailed mechanisms of GRI 2.11, GRI 3.0, and San Diego has been studied under the laminar flamelet modelling framework. Computed results have been compared against the well-known experimental data of Sydney University bluff-body CH 4 /H 2 flame. Results show that the laminar flamelet model yields very good agreement with measurements for temperature and major species with all the reaction mechanisms. The GRI 2.11 performs better than the other reaction mechanisms in predicting minor species such as OH and pollutant NO. The agreement achieved for NO is particularly encouraging considering the simplified modelling formulation utilized for the kinetically controlled NO formation.

A methodology of extending laminar flamelet model in its adiabatic form to a nonadiabatic form which can account for radiative heat loss as well as its effect on NO x pollutant has been developed. Coupling of radiation submodel with flamelet model is based on the enthalpy defect concept. Pollutant NO x has been calculated from solution of its transport equation containing source term which is derived from flamelet calculations. Flamelet calculations adopted GRI 2.11 reaction mechanism which accounts for detailed carbon and NO x chemistry. Depending on consideration of variation in scalar dissipation within flamelet calculations, the non-adiabatic form has been further divided into non-adiabatic model with single (NADS) and multiple scalar dissipation rates (NADM). Bluff-body stabilized CH 4 /H 2 flame has been chosen as the test case to assess the capability of non-adiabatic models. Turbulence closure has been achieved with a Reynolds stress transport model. Calculations have also been carried out with a modified k-ε model for evaluation of relative performance of the two turbulence closures. Performance of non-adiabatic flamelet models in regard to the overall structure of the flame is reasonably good and the agreement is similar to that of the adiabatic flamelet model thereby indicating weakly radiating nature of the flame. However, the NADM model results in minor but encouraging improvement in NO mass fraction predictions by reducing the extent of overprediction observed with the adiabatic model. In contrast, the NADS model results in overprediction over and above the adiabatic predictions thereby showing that, it is imperative to consider variation in scalar dissipation rate in flamelet calculations to capture the effect of radiation on NO. The results also show that employing the Flow Turbulence Combust (

The prometheus combustor design system aims to reduce the complexity of evaluating combustor designs by automatically defining preprocessing, simulation, and postprocessing tasks based on the automatic identification of combustor features within the computer-aided design (CAD) environment. This system enables best practice to be codified and topological changes to a combustor's design to be more easily considered within an automated design process. The following paper presents the prometheus combustor design system and its application to the multiobjective isothermal optimization of a combustor prediffuser and the multifidelity isothermal optimization of a fuel injector feed arm in combination with a surrogate modeling strategy accelerated via a high-performance graphical processing unit (GPU).

Turbine subsystem cooling design depends on the profile of the non-dimensional temperature distribution function (TDF), measured at a traverse plane upstream of the nozzle guide vane (NGV). To date, the compressor discharge OGV profile was thought to have an insignificant effect on the resulting combustor exit traverse, hence a generic OGV geometry has been used for such tests, which typically remained unchanged between varying combustor designs. The present study however shows that the wake profile of the OGV has a significant influence on the measured combustor exit traverse profile. Experiments were performed at Loughborough University with varying OGV geometries to simulate the aerodynamic field surrounding the combustor. Corresponding numerical analyses were performed using an in-house combustion analysis code with a passive scalar technique to model the CO 2 tracer gas injection and mixing. Analysis of the experimental and numerical simulations confirm that the pressure and velocity profiles presented to the system by an axial flow compressor influence both the mass flow and pressure distributions within the combustor feed annuli. This in turn affects the ratio of the mass flow rates entering the flame tube through the dilutions ports located around the inner and outer annuli. The flow through these ports controls the bulk mixing within the flame tube, resulting in a change in mixture concentration profile measured near the combustor exit. Hence, reproducing engine-representative OGV wake structures for a given engine together with an accurate representation of the combustor configuration is of key importance to reproducing the temperature profiles that inform turbine cooling design. NOMENCLATURE CP Chemically pure FSN Fuel spray nozzle ṁ mass flow rate (kg/s) NGV Nozzle guide vane OGV Outlet guide vane ppm parts per million P Pressure (N/m 2 ) RIDN Rear inner discharge nozzle RODN Rear outer discharge nozzle T Temperature (K) TDF Temperature distribution function U Velocity (m/s)