Data Dependent Systems Approach to Modal Analysis Via State Space

Abstract

The application of the Data Dependent Systems (DDS) methodology is proposed for the modal identification and characterization. Difference equation models derived from sampled free and forced response information are shown to quantitatively define the dynamic properties of the system. A modal decomposition of the DDS models, based on modern linear system theory in the state space format and analysed from the standpoint of theoretical modal analysis, yields the system eigenvalues and eigenvectors, and the parameters of a lumped structural model. The theoretical background is presented in detail; its validity is established by means of a simulation study. Advantages of the DDS approach are discussed. Although the simulation study, and to some extent the terminology, are based on mechanical systems, the method has applications in wider fields, e.g., electromagnetic radiation and scattering.