Multiple pure tone noise prediction

This paper presents a fully numerical method for predicting multiple pure tones, also known as “Buzzsaw” noise. It consists of three steps that account for noise source generation, nonlinear acoustic propagation with hard as well as lined walls inside the nacelle, and linear acoustic propagation outside the engine. Noise generation is modeled by steady, part-annulus computational fluid dynamics (CFD) simulations. A linear superposition algorithm is used to construct full-annulus shock/pressure pattern just upstream of the fan from part-annulus CFD results. Nonlinear wave propagation is carried out inside the duct using a pseudo-two-dimensional solution of Burgers? equation. Scattering from nacelle lip as well as radiation to farfield is performed using the commercial solver ACTRAN/TM. The proposed prediction process is verified by comparing against full-annulus CFD simulations as well as against static engine test data for a typical high bypass ratio aircraft engine with hardwall as well as lined inlets. Comparisons are drawn against nacelle unsteady pressure transducer measurements at two axial locations as well as against near- and far-field microphone array measurements outside the duct.     

Correction of fan noise for effects of forward flight

When acoustic measurements are made on a static engine test stand, the data must be corrected for the effects of forward flight to predict correctly the noise characteristics of the engine in flight. A ray tracing approach is used here to relate the static test case to the flight case. The assumptions of isentropic irrotational flow into the fan inlet and a cylindrical shear layer at the fan exhaust lead to slightly different methods for correcting inlet noise and exhaust noise. The forward flight correction method generally involves both an angle and an amplitude correction. The amplitude correction factors for inlet and exhaust noise are the same as that for a dipole and can be as much as 6 dB for a flight Mach number of 0·3. The angle correction for the inlet noise differs from that of the exhaust noise, and both differ from the generally used correction to retarded angle.     

Noise simulation of aircraft engine fans by the boundary element method

Numerical simulation results of the civil aircraft engine fan stage noise in the far field are presented. Non-steady-state rotor–stator interaction is calculated the commercial software that solves the Navier–Stokes equations using differentturbulence models. Noise propagation to the far acoustic field is calculated by the boundary element method using acoustic Lighthill analogies without taking into account the mean current in the air inlet duct. The calculated sound pressure levels at points 50 m from the engine are presented, and the directional patterns of the acoustic radiation are shown. The use of the eddy resolving turbulence model to calculate rotor–stator interaction increases the accuracy in predicting fan stage noise.     

Experimental source characterization techniques for studying the acoustic properties of perforates under high level acoustic excitation

This paper discusses experimental techniques for obtaining the acoustic properties of in-duct samples with non-linear acoustic characteristic. The methods developed are intended both for studies of non-linear energy transfer to higher harmonics for samples only accessible from one side such as wall treatment in aircraft engine ducts or automotive exhaust systems and for samples accessible from both sides such as perforates or other top sheets. When harmonic sound waves are incident on the sample nonlinear energy transfer results in sound generation at higher harmonics at the sample (perforate) surface. The idea is that these sources can be characterized using linear system identification techniques similar to one-port or two-port techniques which are traditionally used for obtaining source data for in-duct sources such as IC-engines or fans. The starting point will be so called polyharmonic distortion modeling which is used for characterization of nonlinear properties of microwave systems. It will be shown how acoustic source data models can be expressed using this theory. Source models of different complexity are developed and experimentally tested. The results of the experimental tests show that these techniques can give results which are useful for understanding non-linear energy transfer to higher harmonics.     

The effect of turbulence damping on acoustic wave propagation in tubes

The attenuation of sound due to the interaction between a low Mach number turbulent boundary layer and acoustic waves can be significant at low frequencies or in narrow tubes. In a recent publication by the present authors the acoustics of charge air coolers for passenger cars has been identified as an interesting application where turbulence attenuation can be of importance. Favourable low-frequency damping has been observed that could be used for control of the in-duct sound that is created by the engine gas exchange process. Analytical frequency-dependent models for the eddy viscosity that controls the momentum and thermal boundary layers are available but are restricted to thin acoustic boundary layers. For cases with cross-sections of a few millimetres a model based on thin acoustic boundary layers will not be applicable in the frequency range of interest.In the present paper a frequency-dependent axis-symmetric numerical model for interaction between turbulence and acoustic waves is proposed. A finite element scheme is used to formulate the time harmonic linearized convective equations for conservation of mass, momentum and energy into one coupled system of equations. The turbulence is introduced with a linear model for the eddy viscosity that is added to the shear viscosity. The proposed model is validated by comparison with experimental data from the literature.     

Computational aero-acoustics for fan duct propagation and radiation. Current status and application to turbofan liner optimisation

Novel techniques are presented to reduce noise from turbofan aircraft engines by optimising the acoustic treatment in engine ducts. The application of Computational Aero-Acoustics (CAA) to predict acoustic propagation and absorption in turbofan ducts is reviewed and a critical assessment of performance indicates that validated and accurate techniques are now available for realistic engine predictions. A procedure for integrating CAA methods with state of the art optimisation techniques is proposed in the remainder of the article. This is achieved by embedding advanced computational methods for noise prediction within automated and semi-automated optimisation schemes. Two different strategies are described and applied to realistic nacelle geometries and fan sources to demonstrate the feasibility of this approach for industry scale problems.     

开孔结构流致噪声的数值模拟和机理分析

飞机机体表面的开孔设计会形成空腔结构,产生空腔流致噪声。空腔噪声的控制需要彻底认识其流动和噪声机理。以飞机的功能性开孔为例,通过半经验公式分析了其空腔噪声频率随速度的变化规律,预测了出现流声共振的工况。空腔发生流声共振时,特定频率的纯音噪声会被放大。为此,采用脱体涡模拟方法开展了开孔结构流声共振的三维非定常数值计算,分析了其流场和声场特性。其中,数值方法的准确性通过圆形空腔标模计算进行验证。结果表明,在一定速度下剪切层内的扰动将诱发空腔深度方向声模态,出现流声共振现象。此时,剪切层表现为强烈的周期性上下拍动,空腔底部和后缘区域的局部压力脉动幅值较大,声波主要由空腔后缘向上游方向辐射,上游噪声大于下游。     

“Buzz-saw” noise: Prediction of the rotor-alone pressure field

Public expectations of lower environmental noise levels, and increasingly stringent legislative limits on aircraft noise, result in noise being a critical technical issue in the development of jet engines. Noise at take-off, when the engines are at high-power operating conditions, is a key reference level for engine noise certification. “Buzz-saw” noise is the dominant fan tone noise from modern high-bypass-ratio turbofan aircraft engines during take-off. Rotor-alone tones are the key component of buzz-saw noise. The rotor-alone pressure field is cut-off at subsonic fan tip speeds; buzz-saw noise is associated with supersonic fan tip speeds, or equivalently, high power engine operating conditions. A recent series of papers has described new work concerning the prediction of buzz-saw noise. The prediction method is based on modelling the nonlinear propagation of one-dimensional sawtooth waveforms. A sawtooth waveform is a simplified representation of the rotor-alone pressure field. Previous validation of the prediction method focussed entirely on reproducing the spectral characteristics of buzz-saw noise; this was dictated at that time by the availability of spectral data only for comparison between measurement and prediction. In this paper, full validation of the method by comparing measurement and prediction of the rotor-alone pressure field is published for the first time. It is shown that results from the modelling based on a one-dimensional sawtooth waveform capture the essential features of the rotor-alone pressure field as it propagates upstream inside a hard-walled inlet duct. This verifies that predictions of the buzz-saw noise spectrum, which are in good agreement with the measured data, are based on a model which reproduces the key physics of the noise generation process. Validation results for the rotor-alone pressure field in an acoustically lined inlet duct are also shown. Comparisons of the measured and predicted rotor-alone pressure field are more difficult to interpret because the acoustic lining significantly modifies the sawtooth waveform, but there remains good agreement with the measured spectral data. The buzz-saw noise prediction code used to generate the simulations in this paper has been used by the Rolls–Royce Noise Department since 2004.     

Estimation of the far-field directivity of broadband aeroengine fan noise using an in-duct axial microphone array

This paper presents a measurement technique for estimating the far-field directivity of the sound radiated from a duct using measurements of acoustic pressure made inside the duct. The technique is restricted to broadband, multi-mode sound fields whose directivity patterns are axi-symmetric, and whose modes are mutually uncorrelated. The technique uses a transfer function to relate the output from an in-duct axial beamformer to measurements of the far-field polar directivity. A transfer function for a hollow cylindrical duct with no flow is derived, and investigated in detail. Transfer functions for practical cases concerning aeroengine exhausts are also presented. The transfer function is shown to be insensitive to the mode-amplitude distribution inside the duct, and hence can be used to predict the directivity in practice where the noise source distribution is unknown. The technique is then validated using a no-flow facility, and is shown to be able to predict variations in the far-field directivity pattern and also estimate the far-field sound pressure levels to within 2 dB. It is suggested that the proposed technique will be especially useful for fan rig experiments, where direct measurement of directivity, for example by use of an anechoic chamber, is impossible.     

Reduction of fan noise by annulus boundary layer removal

Much of the noise produced by a fan is due to varying forces caused by the unsteady flow field through which it passes. In the absence of inlet guide vanes, support struts and other mechanical obstructions, the flow irregularities are caused by large scale intake turbulence and temporal and spatial variations in the annulus boundary layer. In many cases the most effective source is the annulus boundary layer, as this interacts with the blade tip which is the fastest part of the blade and is therefore the most effective noise source. This noise can be reduced if the annulus boundary layer is smoothly bled away from the rotor so that the blade tip sees a thin, uniform boundary layer. A one metre diameter ventilation fan has been run in a duct with and without a boundary layer bleed system and a reduction of far field sound power of approximately 5 dB has been obtained for a bleed of 5 % of the main flow with some far field tone reductions of more than 15 dB. Measurements of far field directivities, in-duct acoustic modes and aerodynamic distortions have confirmed the suggested explanation for this phenomenon. The tests have shown that great care must be taken in the design of a bleed system so that residual boundary layer distortions are not in the correct wavelength range to produce propagating acoustic waves at important blade passing harmonic frequencies. If such distortions are present, the far field noise can increase instead of decrease.     

Effectiveness of combined aircraft engine noise suppressors

We consider the design features of fan noise suppressors in application to air intakes and the bypass duct of a turbofan engine. A combined liner is developed that has increased acoustic efficiency in comparison to conventional honeycomb liner. We demonstrate the important role of the area of the sound-absorbing liner between fan Rotor and Stator ensuring significant noise reduction.     

Prediction of flow-generated noise produced by acoustic and aerodynamic interactions of multiple in-duct elements

Flow-generated noise problem caused by in-duct elements is due to the complicated acoustic and turbulent interactions of multiple in-duct flow noise sources. The approach of partially coherent sound fields used previously by Mak and Yang [C.M. Mak, J. Yang, Flow-generated noise radiated by the interaction of two strip spoilers in a low speed flow ducts, Acta Acust united with Acustica 88 (2002) 861–868] and Mak [C.M. Mak, A prediction method for aerodynamic sound produced by multiple elements in air ducts, J Sound Vib 287 (2005) 395–403] is adopted to formulate the sound powers produced by interactions of multiple elements at frequencies below and above the cut-on frequency of the lowest transverse duct mode. The study indicates that the level and spectral distribution of the additional acoustic energy produced by the interactions of multiple elements can be predicted based on the measured data with respect to the interactions. The proposed method can form a basis of a generalized prediction method for flow-generated noise produced by multiple elements. The application of the proposed method is supported by two engineering examples.     

电子器件散热风扇气动噪声管道声学模态截止控制技术

深度学习输入特征的选择直接影响其分类性能,为了进一步提高基于深度学习的鸟类物种识别模型的分类性能,该文提出一种多特征融合识别方法。该方法首先通过短时傅里叶变换、梅尔倒谱变换和线性调频小波变换分别计算得到鸣声信号的3种语图样本集,然后分别利用3种语图样本集训练3个基于VGG16迁移的单一特征模型,将3个模型的输出进行自适应加权求和实现融合,并修正了加权交叉熵函数以克服样本不平衡的问题,最后对语图进行分类实现鸟类物种的识别。以ICML4B鸣声库的35种鸟类为研究对象,对比了4种模型的平均识别准确率(MAP),结果表明特征融合模型较单一特征模型的MAP最大提高了0.307;选择输入语图的持续时间分别为100 ms、300 ms以及500 ms,对比不同持续时间下4种模型的测试MAP值,结果表明持续时间为300 ms时4种模型的MAP值均为最高;对比了不同信噪比下4种模型的识别效果,多特征融合模型的识别准确率随着信噪比的下降降低最少。说明在选择合适的语图持续时间后,该文提出的特征融合模型能得到更高的识别准确率,具有一定的抗噪能力,且训练参数少,更适合于少样本鸟类的识别。     

Role of structural noise in aircraft pressure cockpit from vibration action of new-generation engines

The evolution of new-generation aircraft engines is transitioning from a bypass ratio of 4–6 to an increased ratio of 8–12. This is leading to substantial broadening of the vibration spectrum of engines with a shift to the low-frequency range due to decreased rotation speed of the fan rotor, in turn requiring new solutions to decrease structural noise from engine vibrations to ensure comfort in the cockpits and cabins of aircraft.     

Slat noise modeling and prediction

This paper presents a model for aircraft slat noise prediction, based on the theory of aerodynamic sound generation and the first principles of source flow physics. Starting from the theory of acoustic analogy, the noise from the high Reynolds number and low Mach number flows in the slat cove region is formulated as a general solution by the method of dimensional analysis, far-field asymptotic expansion and statistical modeling. The solution relates the far-field noise spectrum to the surface pressure statistics, the characteristic length and time scales in the surface pressure statistics and Green's function that accounts for the sound-flow coupling and propagation effects. The general solution is then used to extract scaling laws and correlation models for the individual functional dependences between the far-field noise and various parameters, including the slat noise spectral shape, its Mach number dependence and its far-field directivity. The simple scaling laws and correlation models are validated by test data and serve as building blocks to construct a slat noise prediction model.     

A hybrid approach for aeroacoustic analysis of the engine exhaust system

This paper presents a new hybrid approach for prediction of noise radiation from engine exhaust systems. It couples the time domain analysis of the engine and the frequency domain analysis of the muffler, and has the advantages of both. In this approach, cylinder/cavity is analyzed in the time domain to calculate the exhaust mass flux history at the exhaust valve by means of the method of characteristics, avoiding the tedious procedure of interpolation at every mesh point and solving a number of equations simultaneously at every junction. This is done by making use of an interrelationship between progressive wave variables of the linear acoustic theory and those of the method of characteristics. In this approach, nonlinear propagation in the exhaust pipe is neglected and free expansion is assumed at the radiation end of the exhaust pipe. In the case of a muffler proper, expansion from the exhaust pipe into the first chamber is assumed to be a free expansion. Various results of this approach are compared with those of the method of characteristics and the classical acoustic theory, and various peaks and troughs in insertion loss curves are analytically validated.     

Prediction of intake noise of an automotive engine in run-up condition

It is very important to predict the radiated noise from the engine intake system for the effective noise control and virtual prototyping of in-cavity and outdoor noise of a vehicle. To this end, one should precisely measure the in-duct acoustic source parameters of the intake system, viz., source strength and source impedance. Usually, the noise radiation characteristics need to be expressed as a function of engine speed. In this study, acoustic source parameters of an engine intake system under engine run-up condition were measured by using the direct method. Direct method employed two external loudspeakers, turned on simultaneously, and three microphones for the separation of upstream and downstream wave components. It was noted that the frequency spectra of source impedance hardly changes with the increase of engine speed. Utilizing this fact, source strength under the engine run-up condition was calculated by assuming invariant source impedance. Predicted insertion loss and radiated sound pressure level using the measured source parameters were compared with those of measured data and predicted data using several idealized source models, which have been adopted for the calculations. A reasonably good agreement was observed between measured sound spectra at the intake orifice and predicted one using the measured source data. It was shown that the source data obtained by the present method yielded a far better prediction accuracy than those by the idealized source models.     

Development of a prediction method for flow-generated noise produced by duct elements in ventilation systems

Flow-generated noise generated on the quiet side of the primary attenuators of a ventilation system is the result of interaction between air flow and duct discontinuities. It is of engineering importance to predict the flow-generated noise caused by air duct elements in ventilation systems at the design stage. However, all prediction methods are based upon an isolated in-duct element that is very different from a real ventilation system. Until recently, Mak and Yang have produced a prediction method for flow-generated noise produced by the interaction of two elements in air ducts. In this paper, an attempt has been made to modify their equations so that their predictive equations can possibly be used to predict noise produced by “real” duct discontinuities. By comparing their predictive values with the experimental results of Oldham and Ukpoho, their validity can be proved. The modified Mak-Yang predictive equations, therefore, provide a basis for permitting a more accurate prediction of flow-generated noise produced by various configurations of two in-duct elements and duct dimensions.     

Fan interaction noise reduction using a wake generator: experiments and computational aeroacoustics

A control grid (wake generator) aimed at reducing rotor-stator interaction modes in fan engines when mounted upstream of the rotor has been studied here. This device complements other active noise control systems currently proposed. The compressor model of the instrumented ONERA CERF-rig is used to simulate suitable conditions. The design of the grid is drafted out using semi-empirical models for wake and potential flow, and experimentally achieved. Cylindrical rods are able to generate a spinning mode of the same order and similar level as the interaction mode. Mounting the rods on a rotating ring allows for adjusting the phase of the control mode so that an 8 dB sound pressure level (SPL) reduction at the blade passing frequency is achieved when the two modes are out of phase. Experimental results are assessed by a numerical approach using computational fluid dynamics (CFD). A Reynolds averaged Navier-Stokes 2-D solver, developed at ONERA, is used to provide the unsteady force components on blades and vanes required for acoustics. The loading noise source term of the Ffowcs Williams and Hawkings equation is used to model the interaction noise between the sources, and an original coupling to a boundary element method (BEM) code is realized to take account of the inlet geometry effects on acoustic in-duct propagation. Calculations using the classical analytical the Green function of an infinite annular duct are also addressed. Simple formulations written in the frequency domain and expanded into modes are addressed and used to compute an in-duct interaction mode and to compare with the noise reduction obtained during the tests. A fairly good agreement between predicted and measured SPL is found when the inlet geometry effects are part of the solution (by coupling with the BEM). Furthermore, computed aerodynamic penalties due to the rods are found to be negligible. These results partly validate the computation chain and highlight the potential of the wake generator system proposed.     

Analysis of liner performance using the NASA Langley Research Center Curved Duct Test Rig

The NASA Langley Research Center Curved Duct Test Rig (CDTR) is designed to test aircraft engine nacelle liner samples in an environment approximating that of the engine on a scale that approaches the full scale dimensions of the aft bypass duct. The modal content of the sound in the duct can be determined and the modal content of the sound incident on the liner test section can be controlled. The effect of flow speed, up to Mach 0.5 in the test section, can be investigated. The results reported in this paper come from a study to evaluate the effect of duct configuration on the acoustic performance of single degree of freedom (SDOF) perforate-over-honeycomb liners. Variations of duct configuration include: asymmetric (liner on one side and hard wall opposite) and symmetric (liner on both sides) wall treatment; inlet and exhaust orientation, in which the sound propagates either against or with the flow; and straight and curved (outlet is offset from the inlet by one duct width) flow path. The effect that duct configuration has on the overall acoustic performance is quantified. The redistribution of incident mode content is shown, in particular the mode scatter effect that liner symmetry has on symmetric and asymmetric incident mode shapes. The Curved Duct Test Rig is shown to be a valuable tool for the evaluation of acoustic liner concepts.