Material parameter identification of acoustic polymeric foams via theoretical modeling and experimental measurements.

摘要

Unwanted noise can be detrimental to those exposed: negative effects may include hearing loss, sleep loss, and increased stress. Additionally, acoustic radiation may be damaging to sensitive mechanical and electrical systems because excess vibration and fatigue may be induced. For these reasons, noise reduction is of great interest to engineers. Approaches to noise control may be classified as active or passive. Passive noise control widely employs acoustical treatments with porous materials, known to be effective sound absorbers. The sound absorption coefficient provides a quantitative measure of the acoustic energy absorption for rigidly-backed porous materials. This frequency-dependent quantity may be obtained with acoustic measurement or predicted with empirical, microstructural, or phenomenological models. In this work, a phenomenological model for wave propagation in the acoustic impedance tube is presented. We derive free and forced responses of the coupled system and generate the normal-incidence sound absorption coefficient. The frequency range over which the model is valid is assessed. A direct model prediction of the normal-incidence absorption coefficient using known material properties provides this range. We formulate an optimization problem to obtain the set of material parameters for which the generated absorption profile best fits experimental data for a given foam. A piecewise objective function is presented in which the nonlinear constraints are inherent and the least-squares index is to be minimized. We adopt a non-gradient based search method for solution, and present resulting optimal design variables. We study the effect of the frequency discretization on the optimization outcome. The optimally-predicted parameters range within 0.6 to 11.3 percent error of the known material set.