Identification of confidence distributions for modal parameters in the face of measurement uncertainty

Abstract

While classical methods of modal analysis typically disregard measurement uncertainty and identify a single best-fitting set of modal parameters on the Fourier-transformed frequency response, this contribution presents a novel approach that models measurement uncertainty as an additive possibilistic error. A modified Fourier transform is applied to obtain imprecise frequency response functions, based on which an approach to imprecise modal parameter identification is presented. The applicability of the methodology is demonstrated on a simulated example of a damped harmonic oscillator and a real-world experiment of a wooden guitar soundboard.