Mark H. Dunn

Aeroacoustic Scattering via the Equivalent Source Method

A computational method for predicting the scattering of incident engine noise by an airframe in a uniform flow field has been developed. The method is based on the equations of time harmonic, linearized acoustics and employs the equivalent source method for solving an exterior Helmholtz equation boundary value problem. Details of the governing equation development and numerical solution process are presented. The solution methodology has been implemented in a computer program, called the Fast Scattering Code, which accepts user defined operating conditions and aircraft component geometries and simulates the scattered acoustic pressure and/or velocity fields at user specified locations. The code features fast numerical calculations that can be performed on personal computers or workstations for moderate excitation frequencies. Several aeroacoustics scattering cases involving model nacelle, wing, and airframe components are presented to demonstrate the program capabilities.

Aeroacoustic Noise Prediction Using the Fast Scattering Code

The present paper addresses the latest theoretical developments and enhancements to the The Fast Scattering Code (FSC), a computer program that predicts the scattered acoustic field produced by the interaction of time-harmonic, incident sound from a known noise source with an arbitrary airframe/wing/nacelle assembly. The theoretical formulation of the problem has been extended to include the effects of non-uniform flows. The new formulation is shown to be valid under conditions for which spatial derivatives of the flow variables can be neglected. The capability of the code to simulate the acoustic field generated within a full scale nacelle under realistic operating conditions is demonstrated. The inclusion of parallel computing strategies is featured with a calculation of the asymmetric field surrounding a commercial transport when the nacelles contain acoustic sources spinning in the same direction.

Acoustic Simulations of an Installed Tandem Cylinder Configuration

** An assessment of component installation effects on flow generated, far-field sound for a tandem cylinder configuration tested in the NASA Langley Research Center (LaRC) Quiet Flow Facility (QFF) is presented in this paper. The Fast Scattering Code (FSC) was used to simulate the insertion losses resulting from the interaction of incident sound generated by flow over the cylinders with the QFF apparatus. Two sources of sound were used to model the flow generated noise: 1) a line of point dipoles placed at the axis of the rear cylinder, and 2) CFD/CAA generated unsteady pressures on the cylinder surfaces in free flow. As observed at the microphone locations used in the test, the effects of the QFF components on the acoustic field were small, being on the order of 2 dB for all frequencies considered, when the dipole line source was employed. Installation effects were higher when the time dependent surface pressures were used as noise generators, being approximately 6 to 9 dB for the primary shedding frequency and 3 to 5 dB for the second harmonic. This difference is most likely due to the absence of near field flow non-uniformities in the FSC calculations. Inclusion of these effects through the use of penetrable data surfaces located in the linear flow region should yield better estimates of the incident sound field.

Open Rotor Noise Prediction Methods at NASA Langley- A Technology Review

Open rotors are once again under consideration for propulsion of the future airliners because of their high efficiency. The noise generated by these propulsion systems must meet the stringent noise standards of today to reduce community impact. In this paper we review the open rotor noise prediction methods available at NASA Langley. We discuss three codes called ASSPIN (Advanced Subsonic-Supersonic Propeller Induced Noise), FW - H pds (Ffowcs Williams-Hawkings with penetrable data surface) and the FSC (Fast Scattering Code). The first two codes are in the time domain and the third code is a frequency domain code. The capabilities of these codes and the input data requirements as well as the output data are presented. Plans for further improvements of these codes are discussed. In particular, a method based on equivalent sources is outlined to get rid of spurious signals in the FW - Hpds code.

Engine Liner Optimization Using the Fast Scattering Code

*† Results of a study to optimize the acoustic properties of a commercial transport engine nacelle liner using the Fast Scattering Code (FSC) are presented in this paper. The total radiated acoustic power measured at an arc around the nacelle was used as the objective function. Prior to the optimization process, the soft surface boundary condition implemented into the FSC was validated using an analytical solution for monopole source radiation above an impedance boundary plane of infinite area. The advantages and disadvantages of six well-known optimization methods were assessed during the study. Results are presented for three liner configurations at two frequencies and two flow conditions. The investigation shows that 1) the FSC can accurately predict the effects of acoustic treatment, and 2) coupling the FSC with the optimization methods provides an efficient methodology for engine liner studies. The presented results demonstrate numerically and graphically the potential noise reduction benefits of single and multi-liner configurations with and without flow effects.

Scattering of High Frequency Duct Noise by Full Scale Hybrid Wing Body Configurations

** The development of a noise prediction tool for the simulation of engine installation effects involving high frequency sources and large scale configurations of arbitrary geometry is described in this paper. The program, Fast Scattering Code (FSC) v3.1/v3.2, incorporates an iterative conjugate gradient method (CGM) for fast convergence and parallel programming constructs for multi-processor usage in shared and distributed memory environments. The frequency limitations of the single processor codes FSC v2.0/v3.0 are eliminated by using direct matrix-vector multiplication with minimal memory requirements. The applicability of the code is demonstrated through comparisons with experimental data obtained at the NASA Langley Research Center (LaRC) for a 3% scale model of a blended wing body (BWB) airframe. Although the characteristics of the ducted noise generator used in the tests could not be fully simulated, predicted nacelle installation effects are in very good agreement with the measured data. High frequency calculations for other hybrid wing body designs are also discussed.

Scattering of Simulated Broadband Noise by Conventional and Next Generation Aircraft

An approach for the simulation and scattering of incident broadband noise is presented in this paper. The sound generation methodology uses discrete frequency components of point source radiation that are combined to produce a time-dependent broadband signal. The technique was implemented as a post-processor to an existing noise scattering code, and validated for monochromatic and pink noise using available experimental data obtained for canonical shapes. In general, excellent agreement in noise directivity was observed between simulated and measured scattered signals. Amplitude agreement at low frequencies was fair, substantially improving with increasing frequency. The approach was extended to the generation of ducted multiple source broadband noise. Noise radiating from generic flowthrough nacelles with multiple sound sources was scattered by a conventional tube-and-wing configuration and a HWB, highlighting the noise shielding advantages of the latter concept.

Numerical computations on one-dimensional inverse scattering problems☆

Abstract In this note we present an approximate method to detemine the index of refraction of a dielectric obstacle. For simplicity we treat one-dimensional models of electromagnetic scattering. The governing equations yield a second-order boundary value problem, in which the index of refraction appears as a functional parameter. The availability of reflection coefficients yields an additional initial condition. We approximate the index of refraction by a k th-order spline which can be written as a linear combination of B -splines. For N /2 distinct reflection coefficients, the resulting N /2 initial value problems yield a system of N nonlinear equations in N unknowns which are the coefficients of the B -splines.

Advanced Time Domain Noise Prediction Methods for Open Rotors and Installation Effects

The rotor noise prediction code ASSPIN2 (Advanced Subsonic and Supersonic Propeller Induced Noise), based on Farassat’s time domain solutions of the FWH for thickness and loading noise sources, calculates acoustic pressure time signals at user specified field points for single-rotation and contra-rotation applications. Loading noise predictions with ASSPIN2 require the aerodynamic pressure on the rotor blade surfaces as input. Physically accurate noise predictions are obtained with ASSPIN2 from CFD generated blade pressure input. However, the computational cost of producing blade loads using CFD is excessively high for conducting blade design analysis or systems noise studies. Level 1 (non-CFD) rotor blade steady aerodynamics methods are presented that model the interaction of the rotor blades with the uniform stream from forward flight. Level 1 unsteady loading effects are included for ingestion by the rotor of flow non-uniformities from installed aerostructures. These rotor aerodynamics models are implemented in the computer programs SBAC (Steady Blade Aerodynamics Code) and UBAC (Unsteady Blade Aerodynamics Code) which have been automated to produce ASSPIN2 input. Rotor aerodynamics results, that include the effects of the aircraft wings and rotor disc angle-of-attack, are presented for an eightbladed, SR7 rotor at take-off and cruise conditions. Aerodynamic performance calculations with SBAC compare favorably with those generated using CFD results. Noise directivity studies with ASSPIN2 and SBAC/UBAC input are conducted to illustrate the interplay between the prediction codes and to assess the unsteady loading effects of installed structures on the radiated noise field.

Curved Duct Noise Prediction Using the Fast Scattering Code

Results of a study to validate the Fast Scattering Code (FSC) as a duct noise predictor, including the effects of curvature, finite impedance on the walls, and uniform background flow, are presented in this paper. Infinite duct theory was used to generate the modal content of the sound propagating within the duct. Liner effects were incorporated via a sound absorbing boundary condition on the scattering surfaces. Simulations for a rectangular duct of constant cross-sectional area have been compared to analytical solutions and experimental data. Comparisons with analytical results indicate that the code can properly calculate a given dominant mode for hardwall surfaces. Simulated acoustic behavior in the presence of lined walls (using hardwall duct modes as incident sound) is consistent with expected trends. Duct curvature was found to enhance weaker modes and reduce pressure amplitude. Agreement between simulated and experimental results for a straight duct with hard walls (no flow) was excellent.