Acoustical impedance characterization of liners using a Bayesian approach

A characterization method of acoustic treatment composed by a perforated plate and a honey comb layer is described. Those treatments are generally used as liners for aircraft nacelle and are characterized by their impedance. Empirical models or Kundt's tube experiments are generally used in order to characterize impedance under specific conditions which are not representative of physics in aircraft liners (plane wave instead complex waves, normal incidence instead grazing incidence...). An inverse method using Bayesian approach was developed to define specific parameters describing impedance of treatments under similar functional conditions. For this, an analytical solution is implemented in order to evaluate the posterior probability density function of parameters and an evolutionary Markov Chain Monte Carlo method (eMCMC) is implemented to explore probability density space. The validation of the inverse method is realized on a simulated data and a real case based on experimental data is presented.