Performance of Kumaresan and Tufts Algorithm in Liner Impedance Eduction with Flow

Implementation of the Kumaresan and Tufts algorithm to liner impedance eduction in a duct with shear flow is described. The approach is based on a noncausal model of sound propagation coupled with singular value decomposition to identify the acoustic pressure modes. The performance of the algorithm is evaluated by comparing the educed impedance spectra to that educed by a benchmark method. Results are presented using both simulated and measured data over a range of test frequencies, three mean flow Mach numbers, and six test liner structures. When simulated data are used, the impedance spectra educed is in perfect agreement with the exact impedance spectra. When measured data are used, it is found that 1)the reduced rank approximation to the prediction matrix increases the accuracy of the educed impedance, 2)the algorithm performs well except at the antiresonant and resonant frequencies of the liner, and 3)at high enough Mach number, the effects of the gradients in the mean flow boundary layer need to be included in the impedance eduction model.