Leveraging a data-driven approach to simulate and experimentally validate a MIMO multiphysics vibroacoustic system

Document Type


Publication Date



Department of Mechanical Engineering-Engineering Mechanics


The objective of the present work is to evaluate the performance of a data-driven approach to simulate a multiphysics vibroacoustic field. Of particular interest is the complex propagation of anechoic waves in a finite 1D-structure submerged in water. Experimental data from an underwater beam actuated by two macro-fiber composites (MFCs) serves as the experimental framework for the data-driven model. Next, the Vector-Fitting (VF) algorithm estimates a state-space model of the multiphysics dynamics with voltage signals to the two MFCs as inputs, and the structural and the acoustic responses as the outputs. The multi-input–multi-output data-driven model is then used to determine the parameters that result in the optimal anechoic wave, a process that was carried out previously by experimental iterations. The optimal time-domain simulations are validated with experimental results.

Publication Title

Mechanical Systems and Signal Processing