Investigation Into the Effect of Mode Shape Errors on Validation Experiments for Experimental-Analytical Substructuring

Experimental-to-Analytical Substructuring has been investigated as a way to avoid potentially costly tests and analysis on large systems by performing smaller, subcomponent-level tests. However, this has proven more difficult to implement than expected because the substructuring calculations can be sensitive to modal truncation and experimental noise and require the measurement of rotational motions at the interface. These issues seem to have led to decreased confidence in substructuring calculations, causing it to remain underutilized in industry. In recent work, the authors have proposed performing a small, inexpensive validation test to give confidence or reveal flaws in an experimentally-derived model. This work explores whether such an approach might be used to characterize the uncertainty in a substructure model due to measurement errors. Substructuring is simulated on a finite element model of a beam structure whose modal parameters have been contaminated with noise to simulate measurement errors and cross axis sensitivity. The beam is then coupled to a validation fixture, and the modal parameters of the validation experiment are investigated in a Monte-Carlo experiment to determine the correlation of a validation test to the actual application for different levels and types of errors in the measured mode shapes.