Innovative PERM injection system design within the NEWAC EC Project

CFD aerodynamic study and MOGA optimization of the air distribution layout for a medium pressure Combustor

Abbreviations
ACARE: Advisory Council for Aeronautical Research in Europe
ANTLE: Affordable Near-Term Low Emissions
CAEP: Committee for Aviation Environment Protection
CAE: Computer-Aided Engineering
CFD: Computational Fluid-Dynamics
CLEAN: Component vaLidator for ENvironmentally-friendly Aero-Engine
EEFAE: Efficient, Environmentally Friendly Aero-Engine
ESTECO: EnginSoft TECnologie per l’Ottimizzazione
MOGA: Multi Objective Genetic Algorithm
NEWAC: NEW Aero Engine Core concept
OPR: Overall Pressure Ratio
PERM: Partially Evaporating Rapid Mixing
SRA: Strategic Research Agenda
VITAL: enVIronmenTALly Friendly Aero Engine)

Introduction
NEWAC is an initiative from the Engine Industry Management Group that integrates European aero engine manufacturers, the main European aircraft manufacturer (Airbus), small and medium enterprises and industries providing innovative technologies, as well as leading research institutions in the field of aeronautics to provide a step change for low emission engines by introducing new innovative core configurations to strongly reduce CO2 and NOX emissions.

State of the art of Aero Engines
Global air traffic is estimate to grow at an average annual rate of about 5% in the next 20 years. This scenario urgently requires to address environmental penalties: the gases and particles emitted by engines contribute to local air quality degradation in airport vicinities and alter the concentration of greenhouse gases on a global level, leading to climate change. Thus, Europe's aviation industry faces a considerable challenge to satisfy the demand whilst ensuring economic, safe and environmentally friendly air travel.
Large investments have already been made in Europe and the US through R&D programmes and collaborations to reduce the negative environmental effects of aircraft use. In fact, research provides the technologies to improve the performance of existing engine components.

However, even if these technologies permit improvements in emissions, their existing limitations will not allow the industry to reach the goals set in ACARE: to reduce NOX and CO2 emissions and to achieve the ACARE objectives, it is now mandatory to develop new engine configurations and to perform complementary research and development of core engine technologies (high pressure system).

Objectives
ACARE identified the research needs for the aeronautics industry for 2020, as described in the ACARE SRA. Amongst others, the following targets regarding the engine are set, which will be looked for by NEWAC:

• 20% reduction in CO2 emissions per passenger-kilometre whilst keeping specific weight of the engine constant (see Figure 1);
• significant reduction of NOX emissions during the landing and take-off cycle (-80%) and in cruise (-60%) respect to CAEP/2 limit (see Figure 2)

Figure 1: Fuel consumption / CO2 reduction for different core concepts: Newac vs. state of the art

Figure 2: NOX reduction for different core concepts: NEWAC vs. state of the art

The main result of NEWAC will be fully validated, novel technologies enabling a 6% reduction in CO2 emission and a 16% reduction in NOX according to landing and take-off cycle versus the CAEP/2 limit. These results will be integrated with past and existing EC Projects in the field, notably EEFAE (-11% CO2, -60% NOX), VITAL (-7% CO2) and national programmes, thus CO2 can be reduced up to 20% and NOX close to 80%, hence enabling European manufacturers to attain the ACARE 2020 global targets.

The project will address the particular challenge in delivering these benefits simultaneously: many technological developments based on conventional thermodynamic cycles are driven to high temperature and pressure levels to reduce CO2 whilst compromising NOX emissions.

To avoid this conflict a number of innovative core engine concepts will be investigated and key components will be tested and evaluated. All concepts will be based on single annular combustor architecture that offers the highest potential to keep penalties on weight and associated cost with the introduction of lean low emission combustion technology at acceptable levels. On the other hand, the different operating conditions of the various engine sizes will require the improvement of individual lean burn fuel injection concepts. Starting from these models, each partner works on the definition of a new operating configuration.

In addition to technical objectives, NEWAC will lead to the deployment of the technology by preparing the European engine supply chain, including internal production departments of the NEWAC contractors, through dissemination and training actions. NEWAC will also provide a basis for information to be used for the establishment of future legislation aiming at increasing stringency in NOX and CO2 regulations in the aerospace sector.

Added value of an integrated project
NEWAC conforms to the priorities defined for an integrated project framework by conducting multi-disciplinary research on compressors, combustors, core engines, intercoolers, recuperators, ducting, materials and more generally engine design.

The key benefit of integrating these technologies into one project is that, were these technologies to be developed individually or in separate smaller projects, they would have a very limited benefit at engine level; however, when focused and combined as in the NEWAC integrated project, together they enable new designs of core engines that will provide significant benefits.

EnginSoft’s first year task within NEWAC
For a decade and through its CFD team, EnginSoft is strongly involved in combustion activities. The team is particularly active in research projects funded by the EC which have a focus on low emissions, and thus address the environmental impact due to air traffic, which accounts for 2% of the total global emissions.
In the past, EnginSoft has been involved in activities within the ANTLE, TATEF and CLEAN programme for heat transfer and combustion optimization applications. Such activities were also among the first to employ the novel developed optimization platform modeFRONTIER® for industrial applications.

Thanks to its broad expertise in CAE (process simulation, CFD, optimization of design), EnginSoft is one of the 40 partners of NEWAC. The main role of EnginSoft in NEWAC, as a subtask of the Project, is to bring a contribution to the design of the Ultra Low NOX AVIO Single Annular Combustor, by means of:

• the optimization of an innovative injection system technology called PERM (Partially Evaporating Rapid Mixing), applied to medium overall pressure ratios (20 < OPR < 35): the concept is based on swirler technology development and is addressed to achieve partial evaporation and rapid mixing within the combustor, optimizing the location of the flame and the stability of the lean system;
• the design of a Ultra Low NOX combustor chamber, focusing on the optimization of the architecture;
• the improvement of other critical lean combustion technologies, such as advanced cooling systems, fuel control systems and fuel staging concepts.

Innovative Combustor
The combustion system is the only contributor to NOX emissions. Lean combustion technology operates with an excess of air to significantly lower flame temperatures and consequently significantly reduce NOX formation. Up to 70% of the total combustor air flow has to be premixed with the fuel before entering the reaction zone within the combustor module. Therefore, cooling flow has to be reduced accordingly to provide sufficient air for mixing. Lean combustion comprises the lean direct injection of fuel, premixing with air and at least a partial pre-vaporisation of the fuel before initiating the combustion process.
The optimization of homogeneous fuel-air mixtures is the key to achieve lower flame temperatures and hence lower thermal NOX formation.

However, this homogenization has a strongly adverse effect on combustion lean stability, drastically narrowing the operating and stability range. To overcome these stability drawbacks while maintaining good NOX performance, fuel staging is required: this can be performed by internally staged injectors in a single annular combustor architecture creating a pilot and a main combustion zone downstream of a common fuel injector.

Injection system
The first NEWAC activity carried out by EnginSoft is the investigation of the aerodynamic behavior of the PERM injection system developed by AVIO and tested by University of Karlsruhe.

The device consists of a co-rotating double swirler centripetal injector. The injection system is illustrated in Figure 3. The mixer is composed by 2 swirlers (primary and secondary) with 16 radial channels each.
The purpose of the CFD analysis and experimental tests on injection system is to individuate the swirler working flow function, hence the mass flow required in order to reduce emissions under the available pressurization (depending on engine layout). Moreover, the numerical analysis verifies that the injection system provides good mixing and recirculation for future flame stability.

Figure 3: Injection System

In particular, the aim of this activity is to point out any meaningful difference on the injection system performance, depending on:

• Plenums sensitivity: injection system performance has been compared between a large plenum simulating experimental test rig proposed by AVIO (Figure 4) and a small plenum with annular blockage on the outlet simulating engine rig condition proposed by University of Karlsruhe (Figure 5).
• Transient vs. steady state flow field: velocity and pressure fields generated from an aeronautical engine swirled injection system have a typical non stationary behavior, as many articles and publications demonstrate; hence transient studies on these models are useful to evaluate the approximation when the adopted simulation type is steady-state only.

Figure 4: Avio test ring

Figure 5: University of Karlsruhe test rig

Furthermore, turbulence model sensitivity was addressed comparing standard K-espilon with Prandtl Number modifications with SST formulations in both RANS and URANS mode.

The sensitivity analyses on the pressure plenums showed that geometrical downstream chamber shape strongly affect air distribution (see Figures 6-7): the flow field in the University of Karlsruhe model is strongly canalised (downstream chamber diameter is only two times larger than the injection system diameter), while in the Avio test rig model, the flow develops freely in a constant pressure much wider plenum.

Figure 6: Axial velocity – Avio test rig

Figure 7: Axial velocity – University of Karlsruhe test rig

The recirculation areas/volumes for flame stabilization are completely different, being more open in the Karlsruhe test rig and much narrower, with a stronger axial flow, in the Avio test rig. Moreover, the large plenum (Avio) yields a mass flow ratio between the primary and secondary swirler channels of 1.12, whereas the small one (Karlsruhe) with identical boundary conditions yields nearly an inverted ratio of 0.82. Hence the mixing performance is very sensitive to the pressure plenum geometry and BCS.
Results of steady state analyses are close to time averaged results of unsteady state ones. Although a steady state simulation does not give any frequency information, it can be adopted as good approximation as it allows significant CPU time savings and supplies useful data to evaluate the performance of the injection system.

Figure 8: Combustor chamber sketch

Cowl
In a typical aeronautical engine, the air coming from the compressor is discharged into a pre-diffuser that converts a portion of dynamic pressure to static pressure. Then a diffuser receives the air at the pre-diffuser exit and supplies it to and around an aerodynamically shaped cowl, placed ahead of the injection system.
This cowl usually splits the air into three parts: one is the cowl passage to supply air to the injection system and for dome cooling; the others are feeding the outer and inner annulus passages, where air is introduced in the combustor chamber to cool the walls through liners, break the swirl and constrain the combustion area through the dilution holes while the rest exits through the bleed holes.
Different cowl geometry configurations have been evaluated to find the best shape in terms of obtaining good pressurization levels for the injection system and along inner and outer annulus. The velocity field is also of interest as a bad profile can raise separation and recirculation, resulting in a counterproductive pressure drop.
To simplify this work phase, a preliminary 2D study has been performed to provide a general suggestion of the behavior of the fluid upstream the combustor chamber. For this purpose, it is useful to apply the multi-objective optimization technology modeFRONTIER® by ESTECO.
This tool allows to automatically manage a series of processes acting on input parameters in order to achieve the optimal solution according to imposed constraints and objectives. In this case:

• input geometric parameters (curvature, length, position) define the cowl model to be evaluated;
• parametric meshing, CFD simulations and an automatic post-processing procedure are the processes involved;
• the target pressurization level and air splits on the annulus are the system’s constraints and objectives.

The optimization algorithm is MOGA-II. It is an efficient multi-objective genetic algorithm (MOGA) that uses a smart multi-search elitism. This elitism operator is able to preserve some excellent solutions without bringing premature convergence to local optimal fronts. MOGA-II requires only very few user-provided parameters (such as a number of generations, probability of cross-over, selection and mutation), while several other parameters are internally settled in order to provide robustness and efficiency to the optimizer.

In Figure 9 the PIDO (Process Integration Design Optimization) Logic flow which integrates the parametric ICEM scripting and CFX analysis into a modeFRONTIER® workflow, is shown.

modeFRONTIER® supplies several good candidates, such as those in Figure 10. The last configuration represents the best candidate, since there are no evident separation problems relevant to the cowl edges or injection system (Figure 11).

Figure 10: Cowl 2D study – Model evolution supplied by modeFRONTIER®	Figure 11: Cowl 2D study – Velocity field

The results derived from the 2D study have been applied in a 3D investigation on a simplified periodic sector of the actual annular combustor (the simplification consists of considering no flame tube, no liners and simplified bleed holes at the annulus end).

Several configurations have been taken into account in order to find the best pressurization level feeding the injection system and the air split through inner/outer annulus.

Since 3D effects become fundamental, the attention has been focused on cowl shape refinement. Moreover, a preliminary fuel tube has been introduced into the model. This obstruction deeply affects the velocity field generating a wake in the outer annulus.

From all the analysed configurations, only the best one in terms of optimal pressure and velocity fields has been applied to complete combustor analyses (Figure 12).

Figure 12: Cowl 3D Study

Complete combustor
The final stage of the aerodynamic study deals with a periodic sector of the complete combustor: flame tube, liners and bleed holes (with downstream plenums) are now considered (Figure 13).
Compared to conventional combustors, with chambers now in production the main difference is that up to 70% of the total air flow passes through the injection system, leaving only 30% for the inner/outer annuls and successively the liners, dilution and bleed holes. Hence as the cooling and dilution air strongly diminished, the dilution rows are reduced to just one with no differentiation between primary and secondary dilution holes.

Figure 13: Complete Combustor study

The objective in this stage is to reach the best layout for dilution holes in order to optimize the combustion process. Acting on holes’ diameters and positions, it is possible to change air distribution and dilution flow diffusion: these aspects contribute to create a recirculating region that guarantees flame stability for cooling.
modeFRONTIER® will be useful in future activities within this task to evaluate several diffusion holes’ arrangements to optimize the air flow split and, consequently, the combustion process.
Hence this future activity will have to consider reacting flows thus increasing the complexity of the model, and introducing performance targets, such as NOX, OTDF and RTDF profile constraints.

Conclusions
The NEWAC program has been an important opportunity to develop an innovative methodology based on modeFRONTIER® for application in the aerospace field. With this technique, a large number of virtual prototypes might be evaluated and a selection of the best designs may be made directly within the modeFRONTIER® environment, avoiding a large number of prototype constructions and thus allowing a significant reduction of costs and time.
Future activities to be addressed for the continuation of the project are: injection system evaluation procedure, air flow split balancing optimization, reacting flow combustion performance for some points of the flight envelope.

All the aerodynamic and reacting flow results derived from this work will be useful for future NEWAC activities concerning the development of the innovative Injection System based on the PERM concept and the final optimization of the AVIO Combustor configuration in order to meet the performance targets for pollutant emissions.
For further information about the NEWAC Project, please visit:
www.newac.eu


For any questions on this article, please email the authors:

Ing. Lorenzo Bucchieri
EnginSoft CFD Manager
info@enginsoft.it

Ing. Alessandro Marini
EnginSoft CFD Project Engineer
info@enginsoft.it

Ing. Fabio Turrini
AvioGroup Combustion Manager

Ing. Antonio Peschiulli
AvioGroup Combustion Specialist