Characteristics of unsteady total pressure distortion for a complex aero-engine intake duct

https://doi.org/10.1016/j.ast.2018.04.031Get rights and content

Abstract

Some types of aero-engine intake systems are susceptible to the generation of secondary flows with high levels of total pressure fluctuations. The resulting peak distortion events may exceed the tolerance level of a given engine, leading to handling problems or to compressor surge. Previous work used distortion descriptors for the assessment of intake-engine compatibility to characterise modestly curved intakes where most of the self-generated time-dependent distortion was typically found to be dominated by stochastic events. This work investigates the time-dependent total pressure distortion at the exit of two high off-set diffusing S-duct intakes with the aim of establishing whether this classical approach, or similar, could be applied in these instances. The assessment of joint probability maps for time dependent radial and circumferential distortion metrics demonstrated that local ring-based distortion descriptors are more appropriate to characterise peak events. Extreme Value Theory (EVT) was applied to predict the peak distortion levels that could occur for a test time beyond the experimental data set available. Systematic assessments of model sensitivities to the de-clustering frequency, the number of exceedances and sample time length were used to extend the EVT application to local distortion descriptors and to provide guidelines on its usage.

Introduction

Time variant total-pressure distortion in air intakes first began to be recognized as a significant performance limiting consideration for aircraft engine installations in the 1960s [1]. Issues encountered late in the development cycle of a number of aircraft [2], [3] led to a significant effort to characterize such intake flows as well as to understand engine response. As dynamic instrumentation capability was gradually improved, for both in-flight and wind tunnel settings, a wide range of flow distortion characteristics were identified [4]. Flow distortion can arise from a range of aerodynamic aspects such as of boundary layer ingestion, lip separation, shock induced separation and secondary internal flows. As a consequence, large unsteady perturbations of complex total pressure and swirl distortion flow fields are presented to the engine system. The adverse effect on engine performance and operability can be characterized by stall cells within the first stages of the compression system [5] and in the worst case can lead to engine surge [6]. Furthermore, discrete distortion regions can strongly affect the blade loading, mechanical vibration and fatigue life [7].

The effect of steady state total pressure distortion has received significant attention over the years with the development of distortion indices to evaluate radial and circumferential perturbations at an Aerodynamic Interface Plane (AIP) ahead of the engine fan face. Those indices were developed in order to correlate the flow distortion with the loss in surge margin for a give compressor. Although none of the distortion descriptors is universally used, the Society of Automotive Engineers (SAE) [4] developed a series of indices to characterize the intensity of the loss in total pressure and the shape assimilated to the distortion region at the AIP.

Fully embedded propulsion systems usually have complex convoluted ducts in which unsteady flow distortion arises from local flow separations and strong secondary flows. Although the time averaged AIP flow field statistics for such complex intakes provide an indication of the distortion level for a given operating condition, the complex interaction between secondary flows and separation regions can generate correlated features that are obscured by the overall variations of the distortion [8], [9]. It is well established that steady distortion contributes to the reduction of surge margin for a given propulsion system [4].

Recent experimental work investigated the effect of total pressure distortion for an aero-engine [10]. A linear relationship was demonstrated between the level of total pressure loss at the AIP and the reduction in surge margin for the low pressure compressor. A similar trend was also noted with the Circumferential Distortion Index (CDI) and the 60° based Circumferential Distortion descriptor (DC60). However the flow distortion within the S-duct is highly unsteady with large deviations from the mean flow [8], [11]. The effect of the dynamic component of the distortion and its associated peak value has been known to be responsible for the rise of instabilities within the compressor even though the mean levels were within the acceptable operating limits [6]. It has also been shown that engine response is sensitive to the frequency content of the time variant distortion pattern [12] especially when it is within a few percent of the rotor shaft speed.

The dynamic aspect of the distortion was commonly treated by the use of a probabilistic approach to predict the most likely peak instantaneous distortion level from a limited measurement record. This approach may be used to describe how this level would vary with the amount of time spent at the corresponding flight condition [1], [13]. Borg [14] and Melick et al. [15], developed a way of synthesizing distortion data from limited root mean squared (RMS) pressure measurements, which could greatly reduce instrumentation requirements in early-stage concept screening. This would also, much later, provide a useful basis for deriving time-variant distortion data from Computational Fluid Dynamic (CFD) simulations [16]. These approaches were subsequently extended by Sedlock [17] to estimate the most probable maximum level of RMS total pressure at the AIP. All of these methods required that pressure measurements at the engine face were temporally uncorrelated, and it was recognized from the outset that, where this was not the case, a deterministic empirical approach would be required. The Extreme Value Theory (EVT) initially developed by Jacocks [13] may be used to predict the probability of occurrence of a peak value of a given distortion metric. This method, which fits the peak experimental data with a model, is used in the current analysis to calculate the expected maximum distortion level for a given operating time within confidence intervals [18].

For S-ducts, the distortion at the Aerodynamic Interface Plane (AIP) is mainly due to the secondary flows and local separation regions within the intake. Recent unsteady CFD studies [11] for similar ducts to those evaluated in this work, highlighted common aspects of the unsteady distortion and identified key coherent structures at the AIP. Previous Stereoscopic Particle Image Velocimetry (PIV) and CFD analysis [9], [19] found that the unsteady distortion characteristics of the velocity field were associated with different flow modes. A swirl switching mechanism was identified at the AIP caused by the oscillation of the secondary flows. A vertical perturbation mode was also identified due to the shear layer associated with the local separation within the duct. The unsteady distortion at the AIP is commonly quantified by the computation of instantaneous distortion levels based on a 40 probe measurement rake [4], [20], [21]. However these descriptors were initially developed for steady flow and moderate distortion patterns which can inherently filter some localized peak distortion levels at the AIP. There is a need to assess whether these descriptors can be used to not only assess the peak value of the instantaneous distortion but also to characterize the general dynamic of the flow field at the AIP. Delot et al. [22] measured the unsteady total pressure flow field at the exit of an S-duct with a moderate offset (H/L=0.27, Fig. 1a) for an AIP Mach number (MAIP) of 0.2 with 40 high bandwidth total pressure probes. Spectral analysis revealed that the unsteady pressure measurements of the two total pressure loss regions on each side of the symmetry plane on the lower half of the AIP were strongly correlated and out of phase. The identification of two main frequencies of 110 Hz and 220 Hz led to the hypothesis of associated lateral and longitudinal movements of the streamwise vortices at the AIP. The analysis of the unsteady total pressure field performed by Garnier [23] for a more aggressive S-duct (H/L=0.49, Fig. 1b) also identified a lateral oscillation at a frequency of 200 Hz at the AIP for MAIP=0.2. Significant levels of energy associated with a frequency range of 250–450 Hz were also identified near the centre of the AIP. Data from both of these cases is used in the current analysis.

The overall aim of this paper is to assess circumferential and radial distortion statistics and to investigate if EVT can be used to estimate return values and maximum levels of distortion descriptors for complex S-duct flow fields. This paper also provides a systematic assessment of the EVT method based on a de-clustering process and a convergence study to determine if guidelines in terms of acquisition time are sufficient to capture the overall characteristics of extreme distortion events. The methodology developed through this paper is also applied for the first time on unconventional local ring based distortion descriptors which are more appropriate to evaluate local peak distortion events.

Section snippets

Experimental facility and test case

The investigation presented in this paper processes data from two experiments previously performed at ONERA (Modane) using the suck down intake rig R4. The details of the experimental set up can be found in Delot et al. and Garnier [22], [23]. For this project, two circular cross-section S-ducts were tested. Both S-duct diffusers have similar design parameters except for the vertical offset (H/L). For both duct A and B, the centreline curve is made of two symmetrical arcs of radius Rc and

Time averaged and unsteady flow field analysis

The flow field at the Aerodynamic Interface Plane (AIP) is affected by the flow separation within the duct and the classical secondary flows due to the geometry curvature. For the Duct A configuration, Wellborn [24] performed steady velocity measurements at the AIP using a five hole probe and demonstrated the presence a symmetrical pair of vortices. These vortices were also measured using S-PIV by Zachos et al. [8] for the same configuration. A non-uniform total pressure flow field at the AIP

Conclusions

An assessment of the unsteady total pressure field at the AIP of two S-ducts showed notable levels of unsteady flow distortion which primarily depends on the duct vertical offset with the main unsteadiness up to frequencies of about St=1.5. The mean circumferential distortion descriptors are relatively insensitive to the change in S-duct offset. The temporal variation of ΔPC/P shows the presence of peak distortion events that deviate from the time averaged flow field. The increase in S-duct

Conflict of interest statement

No conflict of interest.

References (30)

  • D. Bowditch et al.

    A survey of inlet/engine distortion compatibility

  • D. Zonars

    Dynamic Characteristics of Engine Inlets

    (May 1972)
  • F. Aulehla

    Intake swirl – a major disturbance parameter in engine/intake compatibility

  • Society of Automotive Engineers, Inlet Total Pressure Distortion Considerations for Gas Turbine Engines

    (2017)
  • D.D. Williams et al.

    Some aspects of inlet/engine flow compatibility

    Aeronaut. J.

    (1973)
  • L.M. Wenzel et al.

    Analysis of Dynamic Inlet Distortion Applied to a Parallel Compressor Model

    (May 1977)
  • W.T. Cousins

    History, philosophy, physics, and future directions of aircraft propulsion system/inlet integration

  • P.K. Zachos et al.

    Flow distortion measurements in convoluted aero engine intakes

    AIAA J.

    (Sep. 2016)
  • D. Gil-Prieto et al.

    Convoluted intake distortion measurements using stereo particle image velocimetry

    AIAA J.

    (2017)
  • R.P.M. Rademakers et al.

    Effects of flow distortions as they occur in S-duct inlets on the performance and stability of a jet engine

    J. Eng. Gas Turbines Power

    (2016)
  • D.G. MacManus et al.

    Complex aeroengine intake ducts and dynamic distortion

    AIAA J.

    (2017)
  • N.C. Bissinger et al.

    Basic principles – gas turbine compatibility – intake aerodynamics aspects

  • J. Jacocks

    Statistical analysis of distortion factors

  • R. Borg

    A Synthesis Inlet Method for Estimating Maximum Instantaneous Distortion Based on Measured Inlet Steady State and RMS Pressures

    (1981)
  • J.R. Melick et al.

    Estimating maximum instantaneous distortion from inlet total pressure RMS measurements

  • Cited by (0)

    View full text