LAUSR

Abstract In substructuring approaches in the field of acoustics, subdomains are mostly characterized by impedance or mobility at their interfaces with other subdomains. This allows handling large systems broken down into smaller, easier to solve subdomains and reaching higher frequency ranges. When a subdomain is characterized by experiments, the impedances are often measured on a dedicated setup aimed at uncoupling the subdomain from the other subdomains. This is often hardly feasible and the setup is insufficiently versatile. In the present article, an inverse method is proposed to experimentally identify the surface impedance of a passive acoustic (or vibro-acoustic) subdomain by means of measurements performed on the whole coupled system. An analytical study illustrates the ability of the approach to identify the impedance matrix of a passive subdomain with accuracy. Nevertheless, as many inverse methods, the system of equations to solve is ill-conditioned. To address this issue, an original approach for tackling the ill-posedness of the problem is proposed. It is based on a statistical analysis of different versions of the matrix to identify. Finally, an experimental validation of the proposed approach is detailed and shows a good agreement when comparing to an available analytical solution.