A comparison of time-reversal and cross-spectral beamforming for localizing experimental rod-airfoil interaction noise sources. (October 2018)

• Record Type: Journal Article
• Authors: Mimani, A.; Fischer, J.; Moreau, D.J.; Doolan, C.J.
• Is Part Of: Mechanical systems and signal processing. Volume 111(2018)
• Page Start: 456
• Page End: 491
• Publication Date: 2018-10
• Subjects: Vortex-turbulence interaction with airfoil -- Tonal noise -- Aeroacoustic time-reversal -- Cross-spectral beamforming -- Flow-induced dipole -- Airfoil self-scattering -- Sink and deconvolution technique; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2018.03.029
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: Different Time-Reversal (TR) implementation methods used for localizing rod-airfoil interaction noise sources. TR implemented using a dipole phase-correction method yields a single focal spot. TR without/with phase-correction and Conventional Beamforming (CB) maps are comparable. TR with facility and airfoil modeling yields the scattered field source location and improves resolution. PTRSL damping and CLEAN-SC deconvolution used to enhance TR and CB maps, respectively. Abstract: This paper compares the results of different implementation methods of Time-Reversal (TR) and Conventional Beamforming (CB) array processing techniques for localizing experimental flow-induced rod-airfoil interaction noise sources. Experiments were conducted in an anechoic wind tunnel for low Mach number cross-flow whereby the far-field acoustic pressure was recorded using two line arrays (LAs) of microphones located above and below the rod-airfoil test-model for a range of flow speeds. TR simulations were carried out for the highest flow speed considered, without and with the rigid-wall modeling of the scattering surfaces which include the experimental facility and the airfoil. The predicted location, resolution and strength of the flow-induced dipole source was noted across different frequency bands wherein it was observed that modeling the airfoil during TR simulation helps to identify location of the scattered field source and simultaneously improves resolution. A Dipole phase-correction method for TR is presented (wherein the scattering surfaces are not modeled) which yields a single focal spot, thereby unambiguously localizing the dipole source. However, the predicted source location and focal-resolution in the Dipole phase-correction TR maps were found to be nearly the same as that obtained by TR without phase-correction and without scatterer modeling. It was shown that the CB maps based on monopole/dipole steering vector formulations were highly comparable to their counterpart TR maps in terms of source characteristics, predicted location, strength and focal-resolution. This demonstrates the equivalence of the TR and CB methods for aeroacoustic source localization when the experimental facility and airfoil were not modeled during TR simulations. Additionally, the Point-Time-Reversal-Sponge-Layer (PTRSL) damping and the deconvolution CLEAN-SC techniques used for enhancing the resolution of TR and dipole CB source maps, respectively, were compared. It was shown that while both methods were equally effective in suppressing side-lobes, the former produced a commensurate reduction in focal spot size whilst the latter yields a nearly constant focal spot size across the frequency bands. Moreover, the simultaneous use of the PTRSL damping and scatterer (in particular, the airfoil) modeling during TR not only yields a further enhanced resolution but also improves the accuracy of the scattered field source location in the low-frequency range.