On the Characterization of Nonlinearities in Assembled Structures

Abstract

This work refines a recently formalized methodology proposed by D.J. Ewins consisting of ten steps for model validation of nonlinear structures. This work details, through a series of experimental studies, that many standard test setup assumptions that are made when performing dynamic testing are invalid and need to be evaluated for each structure. The invalidation of the standard assumptions is due to the presence of nonlinearities, both known and unrecognized in the system. Complicating measurements, many nonlinearities are currently characterized as constant properties instead of variables that exhibit dependency on system hysteresis and actuation amplitude. This study reviews current methods for characterizing nonlinearities and outlines gaps in the approaches. A brief update to the CONCERTO method, based on the accelerance of a system, is derived for characterizing a system’s nonlinearities. Finally, this study ends with an updated methodology for model validation and the ramifications for modeling assemblies with nonlinearities are discussed.

Wave-based analysis of jointed elastic bars: nonlinear periodic response

“…Under harmonic operation this manifests as the well-known nonlinear hysteretic characteristics of the jointed structure. Practically, these structures demonstrate an amplitude-dependent softening (reduction in stiffness/resonant frequency with increasing amplitude) and dampening (increase in the effective damping factor with increasing amplitude) as have been observed in many experimental studies (see, for instance [7][8][9][10]). Mathematical modeling of hysteresis has been an active area of research over the past century (see reviews such as [11,12]).…”

Measurement and identification of the nonlinear dynamics of a jointed structure using full-field data, Part I: Measurement of nonlinear dynamics

“…A key difference between amplitude-controlled and force-controlled experiments is that amplitude-controlled experiments maintain a constant response spectrum, and force-controlled experiments maintain a constant force spectrum. The choice in amplitude-or force-controlled has ramifications for the extraction of the backbone curves, as is discussed further in Smith et al [59]. The employed control strategy can be treated as a closed-loop controller, which keeps the response or force level at the given frequency in an acceptable interval by changing the input current for the amplifier.…”