The experimental synthesis of random pressure fields: Methodology

Abstract

This paper presents a methodology for the off-line reproduction of random pressure fields with given spatial correlation characteristics. The simulation method is first presented, together with an outline of the signal processing techniques required. The design of an experimental setup is then detailed in relation with the nature of the simulated pressure fields. Of particular interest is the laboratory synthesis of three different types of partially correlated random excitations: an acoustic diffuse field, a grazing incident plane wave, and a turbulent boundary layer. The corresponding excitations are generated in a semianechoic chamber over a test panel using a near-field array of 16 loudspeakers driven by a set of optimal signals. The loudspeakers are positioned at a suitable distance above a sufficiently dense grid of microphones close to the simulation surface. The mutually correlated drive signals are determined from both the target correlation properties and the acoustic transfer functions measured between the loudspeakers and the microphones. This approach could provide a cost-effective method of reducing the variation of low frequency sound transmission measurements as well as simulating the propeller-induced noise and the boundary layer noise transmitted through aircraft fuselage structures.

Aeroacoustics research in Europe: The CEAS-ASC report on 2010 highlights

“…In order to evaluate the efficiency of noise reduction treatments, a cost-efficient substitute to in-flight or wind-tunnel measurements would be the laboratory simulation of the response of aircraft sidewalls to the TBL excitation. A methodology has been developed for the laboratory synthesis of random wall-pressure fluctuations with given spatial correlation statistics [41]. Theoretical and experimental results have shown that simulation of the TBL using a near-field array of suitably driven loudspeakers is only feasible below 200 Hz due to the rapid decay of the TBL spanwise correlation length with frequency [42].…”

Measurement Techniques of the Sensitivity Functions to Characterize the Vibration Response of Panels Under Turbulent Boundary Layer Excitation

“…A considerable work has been conducted to experimentally characterize the vibration behavior of panels under TBL excitations without the use of those usual test means. Most of the recently developed methods consist in synthesizing the pressure field induced by a TBL using an array of acoustic sources [1][2][3][4][5][6]. The common issues of those methods are: the number of sources required to reproduce small correlation lengths, which becomes prohibitive at high frequency, and/or the inability to accurately reproduce the induced pressure field outside the acoustic wavenumber domain, which is the predominant domain for subsonic TBL.…”

Source Scanning Technique for Simulating TBL-Induced Vibrations Measurements

“…The practical feasibility of using an array of suitably driven shakers has been studied by Robert [5], while the real-time synthesis of spatially correlated random pressure fields with a near-field array of appropriately-driven loudspeakers was studied by Elliot et al [6] and Maury and Bravo [7,8]. Apart from these works, other semi-experimental techniques have been proposed.…”