Lars Koop

Comparison of Iterative Deconvolution Algorithms for the Mapping of Acoustic Sources

The DAMAS2 algorithm is compared with several other Fourier-based deconvolution approaches. One is the Richardson-Lucy method, which is widely used for the deconvolution of astronomical images. The second is a modified gradient-type NNLS approach, where spectral procedures are implemented to accelerate the computations. Both methods require a computational effort similar to the DAMAS2 algorithm. All three algorithms use an approximate shift invariant point-spread function. %do not take the variation of the %point-spread function into account. %They solve the deconvolution problem only approximately. Furthermore it is described how the DAMAS2 and the Fourier-based NNLS algorithms can be embedded in an outer iteration loop to take the variation of the point-spread function into account. The resulting methods have two nested iterations, which require much more numerical effort than a single DAMAS2 iteration loop. All methods are tested with synthetic data. At first an example with a simple linear array and a small opening angle is considered, where the variation of the point-spread function in the source region is negligible. In this test case the results of the DAMAS2, the modified gradient-type NNLS, and the Richardson-Lucy algorithm are compared. It is shown that these algorithms more or less introduce oscillations in the reconstructed source distribution. A second test case with a planar array and a large opening angle is presented, to demonstrate the influence of a strong variation of the point-spread function. It is shown that the approximate methods lead to distorted results, while the methods with nested iterations give a significantly better reconstruction of the source distribution.

Investigation of the unsteady flow field inside a leading edge slat cove

An experimental investigation of the unsteady flow field inside a slat cove was performed to identify possible sound source generation mechanisms. Tests were carried out on a swept constant chord half-model with a stored chord length of 0.45 m. Experiments were conducted in the 2.0 m x 1.4 m wind tunnel facility of the Technical University of Berlin at an angle of attack of 4°, 8°, 12° and 16° and free stream velocities ranging from 20 m/s up to 35 m/s. The slat cove flow field was investigated by Particle Image Velocimitry (PIV) while the radiated sound was recorded with a single microphone. Both measurements were conducted in a synchronized manner to allow a correlation between the flow field and the sound pressure. An inspection of instantaneous PIV images shows a free shear layer emanating from the slat cusp with discrete vortices further downstream. The shear layer impinges on the inner slat surface and some vortices get trapped inside the recirculation area. Based on a simplified form of the Lighthill-equation sound source terms are calculated. A dipole type sound source distribution is found along the shear layer with decreasing strength further down the shear layer. In addition the vortex shedding frequency of the shear layer vortices is estimated from the instantaneous PIV data. The shedding frequency has a Poisson-like probability distribution with center frequencies ranging from 4 kHz to 10 kHz. The mean shedding frequency scales linearly with the mean velocity of the shear layer. It could not clearly be verified that vortex shedding from the slat cusp is the cause for low frequency broad band noise. For a particular configuration several discrete tones can be observed. It is suspected that this tonal noise originates from the junction between the slat and the fuselage.

Microphone-array processing for wind-tunnel measurements with strong background noise

During the last decade, microphone arrays became a standard tool to study aeroacoustic sources. Nowadays, arrays are frequently applied in fly-over measurements 1 and wind tunnel testing. 2 To improve the resolution of the arrays and to reduce disturbing side lobes in the measured source maps, sophisticated deconvolution methods were proposed 3-8 in recent years. First, these methods were tested using synthetic data or data from experiments with generic configurations. 8 In a next step the methods have to be applied to real experiments. This is done in the present paper. It is tested, if the deconvolution works also under difficult conditions, and if the findings of the simulations using artificial background noise 7 can be confirmed in case of real data.

Reduction of Flap Side-Edge Noise: Passive and Active Flow Control

Microphone array and particle image velocimetry measurements have been performed to investigate the potential of passive and active flow control methods for flap noise reduction. The extended wing flap of the swept constant chord half-model was equipped with either a blowing facility as an active flow control device or with other flap side-edge modifications such as wing tip fences, microtabs and winglets. The flap side edge noise is reduced with the blowing configuration between 2 kHz and 5 kHz. The maximum noise reduction of 15:9 dB is achieved at 2:9 kHz. Varying the diameter of the blowing orifices shows that the noise reduction is governed by the momentum rather than the flow rate of the blowing. The winglets and the suction side fence are most effective in reducing the flap side-edge noise. The reduced level with these configurations is even lower than with the active blowing. The vortex generators and the pressure side fence turn out to be the least effective flow control devices. PIV-measurements at an isolated unswept flap model show that the main vortex on the suction side exhibits a very inhomogeneous vorticity distribution which can be attributed to the unstable shear layer that separates at the lower corner of the flap side-edge and to the negative vorticity production on the suction surface. With blowing the vorticity of the shear layer is concentrated in several small vortices. The distance between those vortices and the solid surface increases with higher blowing momentum and because of this the negative vorticity production due to recirculation is disappears. These distinct modifications of the side-edge flow field explain the noise reduction shown in the aeroacoustic measurements.

Aeroacoustic Measurements of a Scaled Half Model at High Reynolds Numbers

The measurement of airframe noise on small-scale models is well known and common practice in conventional wind tunnels. Since conventional wind tunnels cannot generally achieve full-scale Reynolds numbers, measurements during the development process of modern aircraft are often performed in cryogenic and/or pressurized wind tunnels which are capable of higher Reynolds number flows. Thus, the characteristics of the moving fluid are better adapted to the scale model. At the DLR Institute of Aerodynamics and Flow Technology the microphone array measurement technique was further developed to perform measurements in a cryogenic wind tunnel at temperatures down to 100 K. A microphone array consisting of 144 microphones was designed and constructed for this purpose. In this paper, acoustic array measurements performed in a cryogenic wind tunnel are described for various Reynolds numbers using a 9.24% Dornier-728 half model. Additionally, the background noise of the empty test section was measured within the range of the measurements performed on the Dornier-728 half model. Our results seems to indicate a Reynolds number dependency of the measured sound power for various sources.

Microphone-Array Measurements in a Cryogenic Wind Tunnel

The prediction of full-scale airframe noise based on small-scale model measurements via the phased microphone array technique is well known and in common use in closed test sections. Since conventional wind tunnels cannot generally achieve full-scale Reynolds numbers, measurements are often performed in cryogenic and pressurized wind tunnels which are capable of higher Reynolds number flows. Thus, the characteristics of the moving fluid are adapted to the scale of the model. At the DLR Institute of Aerodynamics and Flow Technology the microphone array measurement technique was further developed to perform measurements in a cryogenic wind tunnel at temperatures down to 100 K. To this end, a microphone array consisting of 144 microphones was designed and constructed. In order to use a microphone array in a cryogenic environment, coming to grips with cold hardiness and ensuring long term stability of the array fairing and the electronic devices, especially the microphones, are the primary challenge. Measurements of the radiated noise from a single rod configuration have been conducted. The results showed very good agreement between theory and measured sound radiation. A Reynolds number dependency of the measured and predicted sound power can be shown.

REDUCTION OF FLAP SIDE EDGE NOISE BY ACTIVE FLOW CONTROL

One dominant airframe noise source is situated at the side edge of the extended wing flap. The objective of the present study is to reduce this noise by blowing air into the flap side-edge vortex to displace or destroy the vortical structure and thus reduce the emission of sound. PIV measurements without blowing yield a rather complicated unsteady vortical structure at the flap side-edge which confirms the assumption of a noise source. This is verified by microphone array measurements. They show that the flap side-edge noise, besides other noise sources, is present over a broad frequency range. The flight parameters such as angle of incidence and slat and flap angles, however, determine which noise source is dominant. PIV measurements with blowing show that the vortical structure can be almost completely dispersed and that the maximum vorticity in the vortex core is reduced. Consequently, a reduction of the flap side‐edge noise can be seen in the microphone array measurement. In addition, the sound pressure level in the acoustic far field is reduced by 3 to 4 dB above 1.25 kHz.

Investigation of the systematic phase mismatch in microphone-array analysis

The paper investigates phase deviations as they occur in aeroacoustic array measurements in open jet wind tunnels. The accuracy of an approximative zero-thickness shear layer model, which is used in practice to predict the phase shift between source and microphones, is investigated numerically. The exact wave propagation through a generic shear layer is calculated by solving the full linearized Euler equations using a finite element method and by means of geometrical acoustics. The results are taken to determine the phase deviations between the simplified model and the exact case. It is demonstrated that the position of the zero-thickness shear layer is important for the accuracy at high frequencies. Additionally measurements are made to analyze the phase fluctuations which are generated by the scattering at turbulent structures in the open-jet shear layer. A procedure is proposed to compensate these phase fluctuations and to improve the array efficiency.

Experimental Investigation of Spectral Broadening of Sound Waves by Wind Tunnel Shear Layers

A detailed experimental investigation of the phenomenon of spectral broadening of monochromatic sound waves by wind tunnel shear layers is presented. The determination of the broadening characteristics as a function of aerodynamic (flow velocity and shear layer characteristics) and acoustic (source and receiver positions, tone frequency) properties is performed in a wind tunnel using two different sound source concepts. The performance of each sound source concept and their equivalence for examining experimentally the effect of spectral broadening is evaluated. The magnitude of the tone broadening by the shear layer was quantified as function of the dimensionless scattering parameter. Furthermore a simple approach for the identification of broadened spectral components of tone spectra is presented and its possibilities and limitations are evaluated in detail.

Simultaneous Particle Image Velocimetry and Microphone Array Measurements on a Rod-Airfoil Configuration

The aeroacoustic sound generation process on a rod-airfoil configuration has been investigated by means of simultaneous particle image velocimetry in the near field and phased-microphone-array measurements in the far field. Up to 20,000 particle image velocimetry snapshots per field of view have been recorded. A coplanar multiplane particle image velocimetry system has been used, providing statistically independent samples of the temporal derivative of the velocity field. Both measurements were conducted in a synchronized manner so as to enable the calculation of the cross-correlation between the acoustic pressure and flow quantities derived from the measured velocity fluctuations. The main idea of the concept used in the study presented here was to use the coefficient matrix obtained from the aforementioned correlation to identify regularities in the near-field fluctuations that are related to the radiated sound field. Additionally, a technique to localize flow structures responsible for the far-field sound by means of the calculation of the near-field intensity has been tested.