PSI - Issue 52

Muhammad Raihan Firdaus et al. / Procedia Structural Integrity 52 (2024) 309–322 M.R. Firdaus et al. / Structural Integrity Procedia 00 (2023) 000–000

313 5

Fig. 3. Full solid modelling of float geometry.

(a) Solid-to-solid (b) Acoustic-to-structure

2. Three-dimensional models

(a) Solid-to-solid (b) Shell-to-shell (c) Membrane-to-membrane (d) Shell-to-solid (e) Acoustic-to-structure

In the submodeling, the model whose solution is interpolated onto the relevant parts of the boundary of the sub model is referred to as the “global” model. Those degrees of freedom at nodes on the submodel boundary whose values are defined by interpolating the solution from the global model are called “driven variables”. Submodeling analysis consist of the following procedures: 1. Running a global analysis and saving the results in the area of the submodel boundary; 2. Defining the total set of driven nodes in the submodel; 3. Defining the time variation of the variables in the submodel analysis by specifying the actual nodes and degrees of freedom to be driven in each step; and 4. Running the submodel analysis using the driven variables to drive the solution. In this research shell-to-solid submodeling is applied. The shell-to-solid submodeling is made out of solid elements that replace a region where the conventional shell elements are used in the global model. Shell-to-solid submodeling requires maximum thickness to be defined in the global model. In this submodeling type ABAQUS expects that all the driven nodes on the submodel belong to solid elements and are driven from a global model region that is entirely made up of shell elements. The required boundary conditions for the multi-stage multi-scale modelling utilizing submodel in ABAQUS are shown in Figure 4 for bulkhead components, and Figure 5 for spar components.