Reverse Engineering Gas Turbine Emission Performance: Applied to an Aircraft Auxiliary Power Unit

Abstract

Environmental and future supply pressures are expected to drive aviation towards alternative fuel sources. However little is available in the literature on aircraft landing-takeoff (LTO) cycle gaseous emissions resulting from the combustion of alternative fuels. Considering the different engine configurations existing in today’s commercial aviation fleet, emission experiments of alternative fuels on all engine types are almost impossible. Modelling may provide a solution but the availability of combustor data (geometry and air split details) in the public domain is limited. A reverse engineering technique is developed to recover the air splits and combustion process in gas turbine engine by a CRN and forward predicting the emissions from the engine exhaust. The model was developed and optimised with a Genetic Algorithm against the Jet A-1 experimental emission data obtained from an APU. Results from the optimised CRN emission predictions closely matched the Jet A-1 gaseous emission data. The modelling technique also successfully demonstrated an ability to predict APU gaseous emission data obtained for Synthetic Paraffinic Kerosene (SPK) (neat and 50–50 blended with Jet A-1) and biodiesel. This technique is expected to enhance the emission databank of aircraft and airside emissions.

Genetic Algorithm optimised Chemical Reactors network: A novel technique for alternative fuels emission prediction

“…calculated and measured emissions data in this respect. A Genetic Algorithm approach [4,5] was selected as the most suitable method for this purpose because it is easy to implement without the need to rearrange the governing equations. This is a self-contained solution for problems of a black box process rather than a competition with methods using CFD.…”

“…The fuel injector design is a slinger system which injects the fuel into the primary combustion zone through a row or two rows of holes from the shaft by taking advantage of centrifugal forces acting on the fuel spray by rotating the shaft to assist fuel atomisation. The CO, NO x and UHC emissions of Jet fuel (JetA-1), synthetic gas to liquid jet fuel (GTL), JetA-1 and GTL 50/50 by volume blend and rapeseed bio-diesel were collected during the campaign [1,4,20]. The APU was started on Jet A-1 and once stabilised, it was switched over to an alternative fuel.…”

“…The control settings of GA are presented in the Table 2. These values were adopted and fine tuned by previous works [17,4,33] …”

“…Therefore, an alternative method to simulate the aviation fleet is required by using publicly available data on specific combustors and engine performance such as the International Civil Aviation Organisation (ICAO) engine emissions data bank [3] or experimental data for specific engines. A simulation method called Genetic Algorithm optimised Chemical Reactors Network (GACRN) [4,5] was developed to tackle the problem in predicting alternative fuels gaseous emissions of Carbon Monoxide (CO), Nitrogen Oxides (NO x ) and Unburned Hydrocarbons (UHC) in aircraft engines. The amounts of these gaseous emissions are regulated by the ICAO during the landing-take off cycles.…”

Genetic Algorithms Optimised Chemical Reactors Network Application: Hydrogen Emissions Performance of a V2527 engine

“…This paper is intended to show an alternative way to analyse one part of this challenge, hydrogen emissions, by using simulation. The Genetic Algorithms optimised Chemical Reactors Network (GACRN) [2] has shown that it is possible to reverse engineer the combustor flow splits and reactor sizes by matching the experimental data and forward predict alternative fuels emissions in an Auxiliary Power Unit (APU). Therefore it was applied to convey the study of NOx emissions in a LandingTakeoff (LTO) cycle of hydrogen as a drop-in fuel in a V2527 engine and as an alternative fuel in a modified V2527 engine.…”