The Wave-Based Substructuring (WBS), as an alternative to the conventional Component Mode Synthesis (CMS) approaches in structural dynamics, is used to gain dynamical properties of an assembly through spatial coupling of two or more sub-components. The WBS has been recently developed in order to speedup the numerical computations while retaining acceptable accuracy of the results of the assembly. The WBS approach involves calculation, modification and selection of a limited set of basis functions that, in the frequency range of interest, properly describe the behavior of the coupling interface between the subcomponents. This limited set of basis functions is then used in a spatial-domain coupling resulting in a much lower number of equations than one would get in the case of the conventional CMS methods. This feature makes the WBS especially appealing in iterative-based numerical calculations like in structural-dynamic modification studies or in optimization. However, as the basis functions are dependent on the full-assembly’s properties, and since these properties are generally changing when there is some modification in one or more sub-components, the issue of robustness of the basis functions is addressed in this paper. The paper also exposes some other issues in the basis-function calculation, namely selection and sensitivity. A new scalar function criterion to asses the quality of synthesis or coupling in general is also proposed. The industrial applicability is demonstrated on an industrial-sized, automotive test case. 1

Abstract This study proposes analytical methods to establish the eigenfunction of continuous system due to substructuring and decoupling of discrete subsystems. The dynamic characteristics of updated continuous system are evaluated by the constraint effect of consistent deformation at the interfaces between two systems. Beginning with the dynamic equation for constrained discrete system, this work estimates the modal eigenmode function for the continuous system due to the addition or deletion of discrete systems. Numerical applications illustrate the validity and applicability of the proposed method.