“…On the other hand, information-transferring approaches often have a separate model for the fine geometric scale, whose response is computed under certain assumptions and transferred to the large-scale model. Numerical homogenization techniques based on the representative volume elements (RVEs) are such approaches that perform numerical material characterization analysis and infuse the material response into the large-scale model [14][15][16].…”
The advancements in additive manufacturing (AM) technology have allowed for the production of geometrically complex parts with customizable designs. This versatility benefits large-scale spaceframe structures, as the individual design of each structural node can be tailored to meet specific mechanical and other functional requirements. To this end, however, the design and analysis of such space-frames with distinct structural nodes needs to be highly automated. A critical aspect in this context is automated integration of the local 3D features into the 1D large-scale models. In the present work, a two-scale modeling approach is developed to improve the design and linear-elastic analysis of space frames with complex additively manufactured nodes. The mechanical characteristics of the 3D nodes are numerically reduced through an automated dimensional reduction process based on the Finite Cell Method (FCM) and substructuring. The reduced stiffness quantities are assembled in the large-scale 1D model which, in turn, enables efficient structural analysis. The response of the 1D model is passed on to the local model, enabling fully resolved 3D linear-elastic analysis. The proposed approach is numerically verified on a simplified beam example. Furthermore, the workflow is demonstrated on a tree canopy structure with additively manufactured nodes with bolted connections. The form of the large-scale structure is found based on the Combinatorial Equilibrium Modeling framework, and the different designs of the local structural nodes are based on generative exploration of the design space. It is demonstrated that the proposed methodology effectively automates the design and analysis of space-frame structures with complex, distinct structural nodes.
“…On the other hand, information-transferring approaches often have a separate model for the fine geometric scale, whose response is computed under certain assumptions and transferred to the large-scale model. Numerical homogenization techniques based on the representative volume elements (RVEs) are such approaches that perform numerical material characterization analysis and infuse the material response into the large-scale model [14][15][16].…”
The advancements in additive manufacturing (AM) technology have allowed for the production of geometrically complex parts with customizable designs. This versatility benefits large-scale spaceframe structures, as the individual design of each structural node can be tailored to meet specific mechanical and other functional requirements. To this end, however, the design and analysis of such space-frames with distinct structural nodes needs to be highly automated. A critical aspect in this context is automated integration of the local 3D features into the 1D large-scale models. In the present work, a two-scale modeling approach is developed to improve the design and linear-elastic analysis of space frames with complex additively manufactured nodes. The mechanical characteristics of the 3D nodes are numerically reduced through an automated dimensional reduction process based on the Finite Cell Method (FCM) and substructuring. The reduced stiffness quantities are assembled in the large-scale 1D model which, in turn, enables efficient structural analysis. The response of the 1D model is passed on to the local model, enabling fully resolved 3D linear-elastic analysis. The proposed approach is numerically verified on a simplified beam example. Furthermore, the workflow is demonstrated on a tree canopy structure with additively manufactured nodes with bolted connections. The form of the large-scale structure is found based on the Combinatorial Equilibrium Modeling framework, and the different designs of the local structural nodes are based on generative exploration of the design space. It is demonstrated that the proposed methodology effectively automates the design and analysis of space-frame structures with complex, distinct structural nodes.
“…Recently multiscale plate element were developed based on the higher‐order computational continua () formulation 5 . Experimental validation of the method for fiber metal laminate (), 6 reinforced concrete beams and plates (), 7 and fiberboard sandwich panels (RVE) 8 is discussed in the cited papers. It should be noted that the RVE approach may be used if two assumptions are satisfied, that is, periodicity and scale separation.…”
We present an adaptation of the multiscale finite element method to the analysis of sandwich beams and plates with complex lattice layers. The proposed modification significantly reduces the number of degrees of freedom (even by four orders) due to the anisotropic higher-order coarse-scale approximation and the novel shape functions that take into account the microscale boundary conditions. Moreover, the local iterative corrector scheme Nguyen and Schillinger (2019) adapted for the bending-dominated responses of sandwich structures provides converges of the coarse mesh approximation to the best possible fine-mesh solution. Several numerical examples are presented to demonstrate the capabilities of the method. We found that the proposed modifications of the shape functions and the higher-order coarse mesh approximation increase the convergence rate. Finally, we validated the proposed model by comparison of the numerical results with experimental ones for a sandwich panel with a dual corrugated high-density fiberboard core. Very good consistency of both results was observed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.