Turbofan duct liner design/optimization: A statistical method

A statistical method for design optimization of acoustic liners for turbofan inlets is developed. This approach is motivated by the observation that the performance of acoustic treatment is dependent on the details of the source. For broadband noise the source is inherently random and for tonal noise in most cases the precise source structure is uncertain. The source is defined by many trials of randomly selected modal powers and modal phases, and with an efficient model of duct propagation, statistical descriptions of realized attenuation are generated with repeatable mean and standard deviation. The method is demonstrated for optimization of acoustic treatment for tonal noise, for broadband noise at specific frequencies, and for spectral attenuation. Comparisons of optimized linings with linings developed in a published experimental program are used to investigate performance predictions.