Computer-aided liner optimization for fan noise propagation and radiation

The main object of this thesis is to investigate acoustic lining in turbofan ducts which have optimal attenuation for a tonal and broadband noise sources. Liner attenuation is necessary in modern turbofan engines. There are several techniques for achieving increased attenuation with acoustic liners. Thisstudy concentrated on two-dimensional and three-dimensional piecewise liner distribution simulated with an advanced acoustic models in which the nacelle geometry is kept unchanged while the liner characteristics are tuned to optimum values. Two computational acoustic prediction codes are used. First a semi-analytic method based on a modal representation of the pressure field to propagate sound in a circular duct is used. Then a more realistic geometry is examined with a spectral/finite element method. The code presented allows the treatment of non-axisymmetric nacelles by combining a standard bi-quadratic approximation in the axial and radial directions, with a spectral representation in the circumferential direction.The aim is to predict how different liner configurations, at various flight conditions, affect the attenuation of sound within an inlet. Two different noise models are used: single-mode and multimode. These represent the two principal fan noise sources: tonal and broadband noise. A robust and consistent optimization procedure based on a Response Surface Model and formal design of experiment methods is built and used. The optimization problems presented are demanding in terms of computer resources, number of variables, optimization convergence and acoustic modeling. In order to allowefficient use of computational resources and effective management of the results the design optimization makes use of Grid computing technology within the Geodise environment. The use of these various techniques in combination makes possible improvements of liner designs with more realistic geometries at modest computational cost. The main feature that emerges from the current study is that it is possible to predict and optimize turbofan liner characteristics in terms of impedance and geometrical distribution to attenuate the noiseemission. One of the main indication drawn from this study is the important reduction in terms of fan sound emission when a piecewise azimuthal liner variation is optimized