In design of components, it is more convenient to analyze the component alone by modeling all the interface loads / displacements as boundary conditions rather than modeling the entire assembly. This process of analyzing at component level is referred as sub model analysis. Most of the assembly level model consists of several components and each component may have different owners. In cases where the entire assembly level model cannot be shared with component owners due to confidentiality, Sub model analysis is the preferred approach. Sub-model analysis also reduces overall model size and in turn the solution time.
Sub modeling is widely used approach in static structural analysis. In dynamic analysis, response of any system is a function of its stiffness, damping and inertia properties along with the applied loads. Sub model will have different inertia, stiffness and damping representations when compared with the assembly level model (also known as full model). The effect of these differences on sub model responses compared to full model needs to be investigated to validate sub modeling approach in dynamic analysis.
In this paper it is shown mathematically and also using finite element analyses that the dynamic response of a component is same between sub-model analysis and full model analysis. In sub-model analysis, dynamic behavior of the remaining system is captured by applying interface loads at component interfaces (with remaining model) which in turn is obtained from full model analysis. Frequency domain finite element model with harmonic excitations is considered to validate the sub modeling approach in dynamic analysis. The observations and conclusions inferred from this frequency domain analysis are also applicable for time domain analysis.
This study is carried out using MSC Nastran.
