Introduction
The submodeling technique allows you to follow up a simulation with a detailed study of one or more local regions of interest, with the results from the original simulation providing the boundary conditions for these local simulations. The original model is referred to as the “global” model and a subsequent detailed model is referred to as a “local” model or a “submodel.” You identify the regions to investigate with submodeling during postprocessing of the global model. A submodel simulation typically goes through approximately the same configuration and stress evolution in the submodel region as the global model but with more detailed and accurate resolution of the results. The driven variables can be degrees of freedom at the nodes in the node-based technique, or they can be components of the stress tensor at the integration points of element faces in the surface-based technique.
Submodeling is often much more efficient than using a globally refined model. The accuracy of submodel results relies on sufficient accuracy of the global results that drive results at the submodel boundary “cuts.” Submodeling leverages Saint-Venant’s principle that the difference between the effects of two different but statically equivalent loads becomes very small at sufficiently large distances from the loading. Judgment is required to position the submodel cuts far enough away from locations where you desire greater accuracy.