COMBUSTOR MODELING IN A 1 D FLOW NETWORK

COMBUSTOR MODELING IN A 1 D FLOW NETWORK TOOL

AGENDA Reasons for a 1 -D flow network gas turbine combustor model? 1 D network model approach – – Advantages Flow distribution modelling Heat transfer process modelling Combustion process modelling Flownex Demonstration Model – Why Flownex? – Methodology – Results

REASONS FOR A 1 -D FLOW NETWORK GAS TURBINE COMBUSTOR MODEL? Preliminary design phase or when considering modifications to existing designs it’s essential to make realistic predictions of – Mass flow splits through the various air admission holes – Total pressure losses – Liner temperatures along the length of the combustor Although powerful, CFD solutions of combustors are specialized, time consuming processes and therefor seldom used during initial sizing of a combustor Network tools allow more accurate initial designs, less time is spent on advanced 3 D simulations and rig tests, thus reducing development time and cost

1 D NETWORK MODEL APPROACH Model of independent sub-flows linked together to model a certain process, that can be described by overlaying a network on the system geometry comprising elements that are linked together by nodes Network models employ conservation governing equations thus • Suitable for incompressible and compressible flow • Allow integrated prediction of 1) Flow distribution 2) Combustion 3) Heat transfer Parameters such as gas temperatures, combustion efficiency, gas emissivity, and correlations to predict film cooling efficiencies can be obtained from the empirical formulations

ADVANTAGES • Capable of modelling complicated geometries effectively without difficulty, while maintaining rapid execution • Several design modifications can be investigated easily • Experimental time can be reduced to only a few verification experiments • Data obtained from 1 D analysis the can be used as as boundary conditions for 3 D numerical models

FLOW DISTRIBUTION MODELLING Predict pressure distribution and mass flow rates throughout a gas turbine combustion system • Elements define actual geometrical features e. g. orifices and duct sections in the domain of interest • Overall governing equations are solved within the nodes • While pressure drop–flow relationship is applied for every element – Typically defined using the Darcy-Weisbach or other exponential or empirically determined correlations.

COMBUSTION PROCESS MODELLING The combustion process has to be accounted for when a 1 -d analysis is conducted • Gas temperature effects fluid density, and therefore the flow distributions and pressure loss • Accurate gas temperature profile required for wall temperature predictions NASA CEA program incorporated into networks allows prediction of adiabatic flame temperature as a function of local air/fuel ratio and the chemical equilibrium product concentrations from reactants

HEAT TRANSFER PROCESS MODELLING Allows prediction of liner wall temperatures by conducting heat balance along liner wall Typically inner wall is heated by convection & radiation from the hot combustion gases & cooled on the outside by the annulus air flow through convection and radiation from the outer liner surface to the casing wall Specialized cooling mechanisms can also be employed – – Convection Film & jet impingement cooling Fluid & surface to surface radiation Conduction

HEAT TRANSFER PROCESS MODELLING Internal radiation Typically the major contributor of heat transferred from the hot gases the liner wall Total emitted radiation has two components • “nonluminous” radiation that emanates from certain heteropolar gases, notably carbon dioxide and water vapor • “luminous” radiation that depends on the number and size of the solid particles (mainly soot) in the flame External radiation Heat transferred from the liner to the casing is usually small compared the external convective heat transfer Approximated by assuming gray surfaces with known emissivities and view factors

HEAT TRANSFER PROCESS MODELLING Convection In the absence of more exact data, it is reasonable to assume that some form of the classical heat-transfer relation for straight pipes (Dittus-Boelter) will hold for conditions inside a liner For external convection Re is based on the hydraulic mean diameter of the annulus air space Conduction 2 D axisymmetric Conduction and cross-conduction components

HEAT TRANSFER PROCESS MODELLING Film cooling Air injected through slots axially along the inner liner wall liner provides protective film of cooling air between wall and hot combustion gases Film cooling correlations • • Levebvre & Ballal Wieghardt

HEAT TRANSFER PROCESS MODELLING Impingement Cooling Impinging jets provide an effective transfer heat and can be positioned to provide extra cooling on liner hot spots Jet impingement correlations Florschuetz

FLOWNEX DEMO MODEL Why Flownex? • Commercial 1 -D thermal fluid network tool that employs the conservation of conservation governing equations – Mass – Momentum – Energy • • User friendly GUI Integrated CEA program calculates chemical equilibrium product concentrations from any set of reactants – ideal combustion model Library that employs industry standard correlations for gas turbine heat transfer Scripting capabilities make tool flexible and user definable Solve time Sensitivity analysis setup Easy to link to 3 rd party FEA, CFD or post processing software

FLOWNEX DEMO MODEL Engine • Rolls Royce T 56 turboprop gas turbine engine • Combustor design in use is old compared to modern-day combustion systems but still widely used • Can-annular type combustion section Gouws, J. J. , R. M. Morris, and J. A. Visser. "Modelling of a Gas Turbine Combustor Using a Network Solver. " South African Journal of Science

FLOWNEX DEMO MODEL Single T 56 combustion chamber and description of hole layout & Operating conditions Gouws, J. J. , R. M. Morris, and J. A. Visser. "Modelling of a Gas Turbine Combustor Using a Network Solver. " South African Journal of Science

FLOWNEX DEMO MODEL Methodology • Set up model to predict pressure distribution and mass flow rate distribution validated against test data • Implement heat transfer to predict gas and liner wall temperatures validated against test data • Once base case is established, investigate effect possible geometry modifications have flow and temperature predictions

FLOWNEX DEMO MODEL General Flownex network layout for the combustion chamber

FLOWNEX DEMO MODEL General Flownex network layout for the combustion chamber

FLOWNEX DEMO MODEL Heat transfer and flow layout for film cooling device

FLOWNEX DEMO MODEL

FLOWNEX DEMO MODEL Heat transfer and flow layout for film cooling device Section 2 (distance 100 mm) Film cooling hole set Ae ~64 mm^2 Liner hot side temperature = 943. 9 K Section 2 (distance 100 mm) Film cooling hole set Ae ~128 mm^2 Liner hot side temperature = 743. 47 K Model execution time ~2 seconds