Towards the Automated Optimisation of Liquid Fuel Injector Design for Gas Turbine Engines

Author(s):  
Adam L. Comer ◽  
R. Stewart Cant

Given the trend towards leaner combustor primary zones and concurrent increases in injector air mass flow rates for emissions reduction, an automated fuel injector optimisation procedure is proposed for a generic aero-engine combustor. The modelling assumptions and the design of the toolset to be applied for the optimisation study, as well as preliminary results from the computational tools, are presented. The proposed configuration will enable the consideration of the following design parameters: the number of swirlers, the swirl number for each swirler, the air mass flow splits between the swirlers, and the fuel mass flow split for multiple prefilming surfaces. Results from the unsteady RANS spray combustion solver available through the OpenFOAM software package are combined with semi-empirical correlations in order to estimate and capture trends in emissions. Pattern factor and susceptibility to thermoacoustic oscillations are assessed directly through the simulation output. Due to computational costs, only the cruise condition is considered for optimisation, and off-design considerations have been limited to their impact on preliminary combustor sizing and design. A multi-fidelity optimisation strategy incorporating a multi-objective Tabu Search algorithm is also presented in light of the nature of the problem and the complexity of the design spaces constructed from CFD results.

2021 ◽  
Vol 20 (2) ◽  
pp. 19-35
Author(s):  
N. I. Gurakov ◽  
I. A. Zubrilin ◽  
M. Hernandez Morales ◽  
D. V. Yakushkin ◽  
A. A. Didenko ◽  
...  

The paper presents the results of studying the flow characteristics of liquid fuel in pressure jet atomizers of small-sized gas turbine engines with nozzle diameters of 0.4-0.6 mm for various operating and design parameters. The study was carried out using experimental measurements, semi-empirical correlations and CFD (computational fluid dynamics) methods. The Euler approach, the volume- of- fluid (VOF) method, was used to model multiphase flows in CFD simulations. Good agreement was obtained between experimental and predicted data on the fuel coefficient and the primary spray cone angle at the nozzle outlet. Besides, the assessment of the applicability of semi-empirical techniques for the nozzle configurations under consideration is given. In the future, the flow characteristics in question (the nozzle flow rate, the fuel film thickness, and the primary spray cone angle) can be used to determine the mean diameter of the droplets (SMD) required to fully determine the boundary conditions of fuel injection when modeling combustion processes in combustion chambers of small-sized gas turbine engines.


Author(s):  
Khaled Yousef ◽  
Ahmed Hegazy ◽  
Abraham Engeda

Abstract Computational Fluid Dynamics (CFD) for air/water-vapor and water-liquid two-phase flow mixing with condensation in a vertical inverted U-tube is presented in this paper. This study is to investigate the flow behaviors and underlying some physical mechanisms encountered in air/water-vapor and water-liquid mixing flow when condensation is considered. Water-liquid flows upward-downward through the inverted U-tube while the air/water-vapor mixture is extracted from a side-tube just after the flow oriented downward. The CFD simulation is carried out for a side air/water-vapor mixture volume fraction (αm) of 0.2–0.7, water-vapor mass fraction (Xv) of 0.1–0.5 in the side air/water-vapor mixture and water-liquid mass flowrate (mw) of 2,4,6, and 8 kg/s. The present results reveal that, at lower air mass flow rate, no significant effect of Xv on the generated static pressure at the inverted U-tube higher part. However, by increasing the air mass flow rates, ma ≥ 0.001 at mw = 2 kg/s, and ma ≥ 0.00125 at mw = 4 kg/s, we can infer that the lowest static pressure can be attained at Xv = 0.1. This may be attributed to the increased vapor and air mass flow rates from the side tube which results in shifting the condensation from the tube highest part due to air accumulation. This leads to increasing the flow pressure and decelerating the water-liquid flow. Raising mw from 2 to 4 kg/s at the same vapor mass ratio results in a lower static pressure due to more condensation of water vapor. The turbulent intensity and kinetic energy starts to drop approximately at ma = 0.002 kg/s, and αm = 0.55–0.76 at mw = 2 kg/s for all Xv values but no noticeable change at mw = 4 kg/s occurs. These findings estimate the operational values of air and water mass flow rates for stable air entrainment from the side-tube. Increasing the air and vapor mass ratio over these values may block the evacuation process and fails the system continuance. Likewise more air entrainment from the side-tube will decelerate the water flow through the inverted U-tube and hence the flow velocity will decrease thereafter. Moreover, this study reveals that the inverted U-tube is able to generate a vacuum pressure down to 55.104 kPa for the present model when vapor condensation is considered. This generated low-pressure helps to vent an engineering system from the non-condensable gases and water vapor that fail its function if these are accumulated with time. Moreover, the water-liquid mass flow rate in the inverted U-tube can be used to sustain the required operating pressure for this system and extract the non-condensable gases with a less energy consuming system. The present CFD model provides a good physical understanding of the flow behavior for air/water-vapor and water-liquid flow for possible future application in the steam power plant.


Author(s):  
Khaled Yousef ◽  
Ahmed Hegazy ◽  
Abraham Engeda

Abstract This paper presents a Computational Fluid Dynamic (CFD) simulation for dry air/water-liquid and two-phase flow mixing in a vertical inverted U-tube using the mixture multiphase and turbulence models. This study is to investigate the flow behaviors and underlying some physical mechanisms encountered in dry air/water-liquid flow in the inverted U-tube. Water flows through the inverted U-tube while the dry air is entrained using the side-tube installed after the water flow downward. The inverted U-tube is tested at water mass flow rates of 2,4,6 and 8 kg/s, air mass flow rates, 0.000614–0.02292 kg/s, with dry air volume fractions 0.2–0.9. The obtained results are compared with the experimental data for model validation and the present CFD model is able to give an acceptable agreement. Also, the results show that, at water mass flow rate of 2 kg/s, there are vortices and turbulent intensity disturbances are noticed at the inverted U-tube higher part, which refers to an air entrainment occurrence from the side-tube. Theses disturbances starts to be stabilized at air mass flow rate around 0.00736 kg/s and air volume fraction, αa = 0.75. This means, if the air mass flow rate increases above this limit, the air entrainment may be blocked. On the other side, at water mass flow rate of 4 kg/s, there are little noticed disturbances until air mass flow rate of 0.00368 kg/s and αa = 0.43 and thereafter stabilized. After this point for water mass flow rate of 4 kg/s, increasing air mass flow rate may block the water flow and the whole inverted U-tube system possible stop flowing. Therefore, this study is able to estimate the required operational conditions and mass ratios for stable air entrainment process. Beyond these operational conditions, air entrainment may be blocked and the whole system discontinues its normal induced gravitational flow. In addition, this study proves that the inverted U-tube is able to generate a vacuum pressure up to 53.382 kPa based on the present geometrical configuration. This generated low-pressure by the inverted U-tube can be used for engineering applications which are working under vacuum and need continuous evacuating form the dry air and non-condensable gases. Furthermore, these findings motivate the utilizing of inverted U-tube for the air evacuation purposes for less power consuming in power plants.


Author(s):  
M. B. Zlatinov ◽  
C. S. Tan ◽  
D. Little ◽  
M. Montgomery

Purge air, injected through seals in the hub of axial turbines, is necessary to prevent hot gas ingestion into endwall cavities, but generates losses by viscous interaction with the mainstream flow. Recent work has shown that for a given purge air mass flow rate, introducing swirl into the purge flow can reduce these losses. This paper investigates the effect of introducing such swirl on the ability of purge flow to prevent ingestion. In particular, it is observed that in the presence of the rotating external pressure non-uniformity due to a downstream blade row, swirled purge flow is much less effective in sealing a turbine disk rim cavity compared to non-swirled purge flow. This is reflected in higher purge air mass flow rates necessary to seal a given cavity, and that in turn diminishes the positive effect of pre-swirling purge flow in the first place. It is shown that this will occur whenever the circumferential pressure disturbance associated with the downstream rotating blades is the dominant driver for externally induced ingestion. It is reasoned that swirled purge flow moves with the rotating pressure non-uniformity and responds to it more readily than non-swirled purge flow, which sees the averaged effect of multiple blade passing events. A flow model based on this physical principle is developed, showing good agreement with computational results. The model yields an ingestion criterion with a parametric dependence on purge flow design parameters. The analysis is extended to an unsteady situation, whereby the effects of both stationary and rotating pressure non-uniformities, from vanes and blades respectively, are taken into account simultaneously. This unsteady flow model points to an optimal design space, in the context of minimizing purge flow losses while maintaining an appropriate margin with regard to hot gas ingestion.


Author(s):  
M. Rabs ◽  
F.-K. Benra ◽  
H. J. Dohmen ◽  
O. Schneider

The present paper gives a contribution to a better understanding of the flow at the rim and in the wheel space of gas turbines. Steady state and time-accurate numerical simulations with a commercial Navier-Stokes solver for a 1.5 stage turbine similar to the model treated in the European Research Project ICAS-GT were conducted. In the framework of a numerical analysis, a validation with experimental results of the test rig at the Technical University of Aachen will be given. In preceding numerical investigations of realistic gas turbine rim cavities with a simplified treatment of the hot gas path (modelling of the main flow path without blades and vanes), so called Kelvin-Helmholtz vortices were found in the area of the gap when using appropriate boundary conditions. The present work shows that these flow instabilities also occur in a 1.5 stage gas turbine model with consideration of the blades and vanes. Therefore, several simulations with different sealing air mass flow rates (CW 7000, 20000, 30000) have been conducted. The results show, that for high sealing air mass flow rates Kelvin-Helmholtz Instabilities are developing. These vortices significantly coin the flow at the rim.


Author(s):  
Racheet Matai ◽  
Savas Yavuzkurt

The performance of an industrial fan was simulated using CFD and results were compared with the experimental data. The fan is used to cool a row of resistor networks which dissipate excess energy generated by regenerative power in an inverter application. It has a diameter of 24 inches (0.6096m) and rotates at different speeds ranging from 2500 to 3900 RPM depending on the requirements. CFD simulation results were also verified by simulating performance of the same fan at different speeds and comparing the results with what was expected from fan affinity laws. The CFD results matched almost exactly (with ∼0.2% difference for pressure at a given flow rate) with the performance being predicted by the affinity laws. The effect of variation of different parameters such as the blade length, number of blades, and blade chord length was studied. Increasing the blade length at the same RPM increased the mass flow rate (by ∼17%) for the same pressure. Increasing the chord length while keeping the same number of blades, at a given RPM, made the performance curve (pressure versus flow rate, i.e. PV curve) steeper and blades stalled at a higher mass flow rate (8.77 kg/sec compared to the previous 8.44 kg/sec). For the same total blade surface area, less number of blades with longer chords stalled at lower mass flow rates (9.22 kg/sec for a 33% shorter chord and 36 blades compared to 8.3 kg/sec for the original rotor which had 24 blades).


Author(s):  
Abdalla M. Al-Amiri ◽  
Montaser M. Zamzam ◽  
Mustapha A. Chaker ◽  
Cyrus B. Meher-Homji

The use of inlet fogging systems to boost the power for gas turbine engines is well known and extensively applied in the power generation field. In this paper the application of inlet fogging of gas turbine engines utilized in the oil and gas sector for mechanical drive applications is covered. Extracting oil from a well is often limited by the rate of gas extraction, and consequently by the gas turbine power and efficiency. In hot and dry air climates, such as desert areas of the gulf countries, gas turbine engine power output is dramatically reduced because of the reduction in gas turbine air mass flow. This effect is even more predominant with aeroderivative units that are commonly used in this sector. Cooling the air to the wet bulb temperature, will increase the density of the air, increase the air mass flow, and boost the power and efficiency. Consequently the amount of extracted gas, and therefore oil, will be substantially increased. With such a cooling potential, and the current trend in oil prices, inlet fogging can have a very rapid payback. In this paper, the behavior of gas turbines with and without fog injection will be analyzed in detail based on actual field data. Critical parameters such as the power turbine inlet temperature, exhaust temperatures, compressor discharge pressure, the gas generator and power turbine speeds, as increasing stages of fogging are applied are covered. Furthermore, specific issues relating to the design and control of fogging as applied to aeroderivative engines will be discussed.


Sign in / Sign up

Export Citation Format

Share Document