Band 39, Heft 1

This paper presents a robust fault identification scheme based on fractional-order integral sliding mode observer (FOISMO) for turbofan engine sensors with uncertainties. The equilibrium manifold expansion (EME) model is introduced due to its simplicity and accuracy for nonlinear system. A fractional-order integral sliding mode observer is designed to reconstruct faults on sensors, in which the fractional-order integral sliding surface guarantees the fast convergence of reconstruction. The observer parameters is selected according to L2 gain theory in order to minimize the effect of uncertainties on the fault reconstruction signal. Simulations in Matlab/Simulink show high reconstruction accuracy of the proposed method despite the present of uncertainties.

Based on the existing compilation technology of load spectrum, this paper proposed a new method for the compilation of the general standard test load spectrum of aircraft engine. On the premise of meeting operating requirements of the actual condition and the basic life test, it simplifies the test load conditions and overcomes the shortage of the existing technology. According to the actual flight of aircraft engine, the load spectrum of a flight mission profile was divided into three parts: the take-off phase, the middle-flight phase, and the landing phase. The proposed compilation method ensures that the generated load spectrum of the take-off phase and landing phase is consistent with the actual flight, and the load spectrum of the middle-flight phase is programmed according to a certain method so that a simple load cycle sequence can be implemented. On that basis, the load spectrum was analyzed and compiled by subsection. The general standard test load spectrum lays the foundation for carrying out the life test and reduces the error between the test and practical use. Since the parameters of the engine load are standardized in the process of compiling, the method can be widely used in various types of engines.

Described in this paper is a numerical investigation on the concept for enhancing film cooling performance by placing vortex generator (VG) upstream the film hole. Six film cooling configurations with different VG locations are investigated, including the distances for VG upstream the film hole (Upstream Distances) of 20 mm, 25 mm, and 30 mm and the VG gaps of 0, 2 mm, 4 mm and 6 mm respectively. The effect of VG locations on film cooling performance were conducted. The film cooling performance is evaluated at the density of 0.97 with the blowing ratio of 1.0. Results obtained show that the twisted flow generated by upstream VG rotates in the opposite direction of kidney vortex, producing an adverse effect on kidney vortex. The enhanced adverse effect could dramatically improve the film cooling performance. Moreover, the intensity of twisted flow is greatly related to the VG locations. The film cooling performance improves with the increasing upstream distances. The film cooling performance improves first and then impairs with increasing VG gaps. In this case, the film cooling performance in the case with gap of 4 mm is superior to those in other cases.

The bleeding air flow increases so as to meet the demand of higher efficiency and better performance of compressor, and the issues brought by the bleeding system is of great interest for more and more researchers. In the opened literature, most of the works were performed through single-passage calculations. However, the non-uniformity caused by the bleeding system lead to an undesirable effect on the compressor, thus the multi-passage calculations should taken into consideration. In the present study, the annulus cascade and bleeding system were numerical studied on ANSYS CFX to investigate the effect of the circumferential position of the bleeding duct holes on the non-uniformity in the compressor and to make it clear that how the non-uniformity is formed. Finally, it is found that the unequal air bleeding mass flow rate in different passages result in the non-uniformity in the compressor. The non-uniformity in circumferential position could be alleviated when the duct hole moving toward the suction surface, however, when the duct hole moving toward the pressure surface, the flow separation would be suppressed in more blade passages.

Successful starting is of utmost importance to any gas turbine engine. In order to ensure successful engine start, the concept of surplus power to accelerate the engine from rest to its idle speed has been proposed in this paper. The concept of surplus power is illustrated with starting data of different types of engines. The effect of variation in surplus power on engine acceleration characteristics has been brought out for in-depth understanding. Analysis of surplus power will provide invaluable information to designers and operators to address issues related to engine starting process. It is found that starting characteristics, i. e. speed vs. time is only a gross characteristic whereas acceleration characteristics dN/dt provides input for minute analysis in surplus power. It is also observed that in case of turbo-shaft engine with free power configuration, turbine is found to come into action much later after light-off whereas in turbo-prop and APU engines with single spool configuration, turbine comes into action immediately after light-off followed by rapid increase in EGT.

A diffusion combustor with a single radial swirler in non-reacting condition is investigated via a large eddy simulation (LES). Three dynamic analysis methods – the fast Fourier transform (FFT), proper orthogonal decomposition (POD) and dynamic mode decomposition (DMD) – are implemented to investigate the flow dynamic characteristics of the combustor. The kerosene-air combustor analyzed in the study was designed by the German Aerospace Center (DLR). It has a square cross-section and uses kerosene as fuel, which is modeled as a pre-vaporized and surrogated fuel consisting of 242 species. The first tangential(1T) mode in combustor caused by the swirler emerges dominantly in the combustor. This 1T mode exhibits the largest amount energy in the combustor dynamics, as verified by POD, and the DMD analysis determines the frequency of 1876.8   Hz. The fuel injector dynamics is associated with Helmholtz resonator frequency of 816.5 Hz. To analyze the instability, the DMD method is employed to investigate the growth rate of the most dominant dynamic structure.

Plasma ignition technology is the research frontier in the field of energy and power. In this paper, a novel type of plasma jet igniter called pre-combustion plasma jet igniter is designed and the experimental results show that it has great potential in strengthening jet stiffness, improving ignition energy, decreasing ignition energy consumption and reducing igniter ablation. The value S called jet stiffness is defined for the first time to reflect the stiffness characteristics of the plasma jet. And the effects of jet gas flow rate, fuel-air ratio and input current on the jet stiffness are investigated experimentally. The experimental results show that the stiffness of the pre-combustion plasma jet is significantly stronger than that of the air plasma jet. When the total jet gas flow rate is increased from 16 L/min to 40 L/min, the stiffness of the air plasma jet and the pre-combustion plasma jet are increased by 73.3 % and 27.2 % respectively. When the current value reaches 30 A, the jet stiffness is increased by about 70 % compared to the input current of 20 A.

Experimental and numerical investigations are carried out on an annular, straight flow, swirl-stabilized aero engine combustor. In this work, the effect of degree and direction of swirl at the inlet of combustion chamber is examined on the liner wall temperature and hot spots. This is carried out by experimentally measuring the liner outer wall temperature at discrete positions along the circumferential and axial directions of the combustor liner in the engine test facility. The RANS based turbulence modeling with reacting flow approach is used to simulate the flow domain. Conjugate heat transfer analysis is used to estimate the liner wall temperature using Ansys CFX frame work. The degree and direction of swirl at the inlet of combustion chamber is found to alter the velocity and temperature profiles inside the combustor and hence found to have a significant effect on the liner hot spots and its location. Hotspot with 43 % increase in temperature near the secondary zone is observed with the increase in swirl angle from 5° to 15° at the combustor inlet. The location of the hotspot is found to be dependent on the swirl direction.

A designed method, multidisciplinary coupling computation and multiobjective optimization, has been established for the composite cooling structure of heavy gas turbine blade manufactured with a directionally solidified Ni-based superalloy. The method combines the one-dimensional fluid network gas-thermal coupling computation, three-dimensional flow field coupled with solid stress field, and anisotropic stress calculation based on finite deformation crystal slip. The temperature, flow field, Von-Mises stress and maximum resolved shear stress of the blade before and after optimization were analyzed. The results show that the optimized blade has lower maximum blade temperature, a more uniform temperature distribution, a lower flow resistance of the coolant channel at the leading edge than that of the original blade. The maximum Von-Mises stress of the optimized blade increases by 10.05 % more than the original blade. The maximum shear stress on the suction side and the pressure surface of the optimized blade are improved and slightly deteriorated compared with that of the original blade, respectively. The corresponding relationship of the maximum shear stress distribution with the local temperature gradient reveals further space for the improvement of the composite cooling structure. This paper has a particular guiding significance for the cooling structure design of the turbine blade.

Stability of flames are affected by fuel properties, geometry of the burner and operating conditions. In this experimental work, first the characteristics of non-premixed flames of Liquefied Petroleum Gas (LPG) and air in cross-flow configuration, where air jet flows perpendicular to the fuel stream, are studied experimentally. Flame transition and stability regimes of non-premixed flames of LPG and air, in a cross-flow burner without and with obstacles, are determined by systematically varying the fuel and air flow rates. Obstacles such as backward facing step and cylindrical bluff bodies are considered. Subsequently, the effects of fuel properties on the stability of flames are analyzed, Flame stability regimes of natural gas (methane) and biogas (methane and carbon-dioxide), measured from a similar burner are available in literature. These have been compared with the stability of LPG flames in terms of power rating of the burner and global equivalence ratio (defined for non-premixed flames).