Band 41, Heft 1

Experimental and numerical study of flame and emission characteristics in a tubular micro gas turbine combustor is reported. Micro gas turbines are used for distributed power (DP) generation using alternative fuels in rural areas. The combustion and emission characteristics from the combustor have to be studied for proper design using different fuel types. In this study methane, representing fossil natural gas, and biogas, a renewable fuel that is a mixture of methane and carbon-dioxide, are used. Primary air flow (with swirl component) and secondary aeration have been varied. Experiments have been conducted to measure the exit temperatures. Turbulent reactive flow model is used to simulate the methane and biogas flames. Numerical results are validated against the experimental data. Parametric studies to reveal the effects of primary flow, secondary flow and swirl have been conducted and results are systematically presented. An analysis of nitric-oxides emission for different fuels and operating conditions has been presented subsequently.

In this study, the flow dynamics with finite volume approach on commercial software Ansys-Fluent 20.0 to solve the compressible two-dimensional fluid flow with Reynolds Average Navier Stokes equation (RANS) equation by considering the density-based solver with Shaer stress transport model (SST) k- ω turbulent model. The species transport model with volumetric reaction and finite rate/eddy dissipation turbulence chemistry interaction is adopted to study the combustion phenomena. Additionally, the effect of spacing between the struts on the flow characters and performance of the combustor is studied by increasing the spacing of struts from 1 mm to 4 mm for each increment of 1 mm. It is found that the multi strut improves the mixing and combustion efficiency compared with that of the single strut owing to the formation of a significant separation layer, resulting in multiple shocks, vortices, and a larger recirculation zone. However, when the spacing of struts is increased further, the performance of the combustor is found to be deteriorating owing to the formation of larger separation layers. The recirculation zone is significant when the strut spacing is minimal and shrinks and restricts itself within the cavity when spacing is increased. So, for better performance of combustor, multi strut with minimum spacing is preferable.

Three-dimensional unsteady numerical simulations were conducted to investigate the detailed film cooling mechanism of the high-pressure turbine shroud with the first-stage turbine blade and guide vane for an aero-engine under the high-speed rotation of blades and rotor-stator interaction. The slip mesh was used to realize the relative motion between the rotating blade and the stationary turbine shroud. It is found that the coolant jet is alternately influenced by the hot mainstream, tip clearance leakage flow and leakage vortex due to the high rotational speed of blades. The film cooling characteristics of the turbine shroud significantly present an unsteady and periodic flow and heat transfer phenomenon. The insufficient cooling margin for film holes at the upstream of the blade leading edge can occur not only under high blowing ratios due to the coolant jet liftoff, but also at low blowing ratios due to the insufficient coolant flow rate as a result of the high exit pressure. A novel shroud cooling structure with coolant supply by the added throttle chamber is put forward, and expected to provide better thermal protection for the high-pressure turbine shroud near the leading edge of blades with no extra increase in the total mass flow rate of coolant.

The high-fidelity finite element model (HFFEM) and Monte Carlo (MC) simulation of the blisk involve large number of calculations, which leads to low computational efficiency. In this case, an improved quasi-static mode compensation method (IQSMCM) and quadratic function-extremum response surface method (QF-ERSM) are proposed to investigate the probability distribution of mistuned blisk based on its vibration characteristics. The number of nodes and elements of IQSMCM relative to HFFEM are, respectively, reduced by 79.66 and 80.03%. Thus, the degrees of freedoms (DOFs) of IQSMCM are obviously reduced compared with that of HFFEM, and its computational efficiency is obviously increased. The maximum displacement shape (MDS) is investigated via IQSMCM. The computational efficiency is enhanced in the condition of ensuring the computational accuracy. Based on the investigation of maximum mode shape, the probability analysis is performed via QF-ERSM. The computational accuracy of QF-ERSM is improved by 93.80% compared with that of MC. Furthermore, the computational efficiency of QF-ERSM is higher 57.06% than that of QF-RSM. The sample history, extremum response surface function, sample history and distribution histogram of MDS are obtained via QF-ERSM, which provides an important guidance for the reliability research of the mistuned blisk. This research can be applied not only to aeroengine’s blisk but also to other large and complex mechanical structures in practical engineering.

A full envelope LMI-based multi-region linear parameter-varying power controller is designed for a turbofan engine in this paper. According to the characteristics of aero-engine model, three scheduling variables are divided into two groups firstly, and then part of them are partitioned, rather than all scheduling variables are partitioned directly as the usual multi-region LPV control. The polynomial LPV model of aero-engine is established under a specific flight condition. An explicit LPV controller by gridding method based on parameter-dependent Lyapunov function is designed and we propose a method to eliminate the dependence of LPV controller on the derivative of scheduling parameter. The flight envelope of turbofan engine is divided into multiple sub-regions, and a mixing LPV control method with overlapping regions is proposed, which can guarantee stability and performance across the full envelope. Finally, the simulation results on the nonlinear component level model of a twin-spool turbofan engine verify our method.

The role of turbine blade cooling and coolants are significant factors in modern gas turbine aerodynamic design. This paper presents an effective and rapid airfoil design method based on CFD computation of the S1 surface and the existing loss correlations. The method can assess the coolant mixing loss by identifying each cooling hole separately and obtain the overall mainflow aerodynamic loss for cooled airfoil. The CFD computation code of the S1 surface is powered by a two-dimensional Euler equation, which is inviscid. Typical Kacker–Okapuu empirical correlations are then used to assess the airfoil friction loss, trailing edge loss, and shock loss. A novel form of the Hartsel model for coolant mixing loss is developed and employed in the CFD codes. In the reformed model, the mixing loss coefficient is directly associated with the blowing ratio and the total pressure/temperature ratio of mainstream-to-coolant, making it more convenient than the original model in the airfoil design process. Based on a transonic turbine vane airfoil, the influences of the film outflow location and outflow Mach number on the coolant mixing loss are investigated using the above prediction method and the cascade blowing test.

Endwall-pulsed blowing (EPB) is studied for three different excitation waveforms to improve the aerodynamic performance of highly loaded compressors. Some important excitation parameters include the excitation frequency and momentum coefficient, which were analyzed in detail. The results of the EPB are compared with the endwall steady blowing (ESB) case. For EPBs with the three excitation waveforms (Waveforms sine, triangle and trapezoid), excitation frequencies that are equal to an integral multiple of the natural frequency of the vortex shedding are optimal and provide better performances than the ESB with the same time-mean momentum coefficient. Moreover, the EPBs of the three excitation waveforms have significant differences in their aerodynamic performance improvements. The optimal case is achieved by the EPB with Waveform triangle and provides a total pressure loss coefficient with a reduction of 25.64%.

In an attempt to reduce engine frontal area, while maintaining a high single stage pressure ratio, mixed flow compressor stages are frequently used in micro gas turbine (MGT) engines. The expansion of the choke margin of such a mixed flow compressor is presented. The use of a crossover diffuser configuration in a mixed flow compressor stage has displayed superior performance results compared to legacy diffuser configurations, especially when geometric restrictions are enforced. A disadvantage of a crossover diffuser configuration is that it typically displays an inferior operating range compared to legacy diffuser configurations. In an attempt to expand the choke margin of a MGT mixed flow compressor, the use of tandem and splitter vane crossover diffuser configurations was evaluated. It was found that a low solidity first vane row configuration provided a 3% increase in choke margin. A splitter vane crossover diffuser configuration provided a 5.9% increase in choke margin. A tandem vaned diffuser with a reduced first row vane number provided a 7.8% increase in choke margin.

The mixing characteristics of a Mach 1.9 jet at three levels of overexpansion, corresponding to nozzle pressure ratio (NPR) 3, 4 and 5, in the presence of a sonic co-flow (secondary flow), which was submerged in a subsonic co-flow (tertiary flow) was studied experimentally. For these NPRs the secondary co-flow is sonic with underexpanded levels and the tertiary flow Mach number was found to be 0.41, 0.71 and 0.85, respectively. The centerline decay results of the primary jet show that the jet mixing is abated by the co-flow, at all levels of expansion. However, in spite of the reduced mixing encountered by the supersonic primary jet, the waves in the jet core are found to be weaker in the presence of co-flows. This may be regarded as an advantage from the shock associated noise point of view, in accordance with Tam’s theory; which states weaker the waves in the core, the lesser is the shock associated noise. The results show that the reduced mixing environment caused by the sonic co-flow alone leads to the jet core elongation of about 20%, 23% and 49%, at NPRs 3, 4 and 5, respectively. The core length of the jet is found to increase by 29%, 46% and 62%, respectively, at NPRs 3, 4 and 5, when both sonic and subsonic co-flow streams are present.

A design for a turbine pre-swirl system with impeller cavity is proposed to improve the quality of cooling air supplied to the turbine blades of an aeroengine. Impeller cavity is analyzed in order to increase the system pressure ratio with a low system power consumption at the system outlet. Theoretical and numerical investigation are used to investigate the flow characteristics in an impeller cavity pre-swirl system. The conclusions in this study indicate that the impeller structure can increases the pressure ratio by changing the power consumption and distribution of the absolute velocity in the impeller cavity and system outlet. To obtain high pressure ratio and low power consumption, the impeller should have a structure with a high outlet installation radius and low outlet angle. The highest increase in the pressure ratio compared with the empty cavity pre-swirl system is 6.4% and the corresponding increase in the power consumption is 2620 W.

This paper numerically studied supersonic film cooling performance in a convergent divergent nozzle, for the purpose of exploring and extending the knowledge of high-efficient cooling techniques for exhaust nozzles in real working conditions. The work was conducted with a steady state RANS approach with the SST turbulence model. The boundary conditions were chosen to mimic a real engine condition. The slot height and the inlet pressure ratio were varied to investigate the effect of the geometrical condition and flow condition. The flow field was analyzed in detail to study the phenomena of the supersonic secondary flow injected into the transonic flow. Besides the cooling effectiveness, the thrust coefficient and the discharge coefficient for the cooling cases were discussed together with the baseline case, to evaluate the influence of the supersonic coolant injection on the nozzle performance. The work can be a basis for the design of cooling schemes in an aero-engine exhaust nozzle.

The ingestion of ice crystals in aero-engine will cause engine surge, flameout, thrust loss, and even in-flight shutdown in the extreme cases, which seriously endanger the flight safety. In order to quantitatively investigate the ice crystal melting characteristic in the compressor, a method based on the compressor mean line flow was developed and validated. Results showed that the wet-bulb temperature increases as the temperature offset increases. The increase in temperature offset or decrease in particle size result in earlier or faster melting of the ice crystals in the low-pressure compressor. The rate of increase in melting ratio decreases with the increase of ice water content at the descent condition. The ambient temperature and ice crystal property are both the important factors affecting the icing risk in the compressor. Higher ambient temperature, smaller particle size or higher ice water content can increase the icing risk in the low-pressure compressor.