Raul Rubio obtained the master’s degree in Aeronautical Engineering by the Universitat Politècnica de Catalunya (UPC, 2018) and he is currently a PhD student in the Composite and Advanced Materials for Multifunctional Structures (CAMMS) group at CIMNE. Under the supervision of professors Àlex Ferrer and Joaquín Hernández, the main area of his research is in the field of preconditioning techniques for iterative solvers. His numerical implementations can be found in an in-house code called Swan https://github.com/SwanLab.
The advancement of technologies such as additive manufacturing, combined with topology optimization techniques, led to the creation of innovative lattice structures with a broad range of applications, from aerospace to the chemical industry. However, solving these structures using the Finite Element Method (FEM) presents significant computational challenges, particularly when dealing with large-scale systems in 3D geometries. Direct solvers struggle with the memory and time demands of these problems, making iterative solvers the only feasible option.
To address the computational burden of numerical simulations, Reduced Order Models (ROMs) were developed to offer faster solutions, though they typically come at the cost of losing some accuracy. In this work, we propose a novel approach by integrating the Empirical Interscale Finite Element Method (EIFEM) into an iterative scheme as preconditioner. This strategy aims to accelerate the iterative solution process while maintaining the accuracy of the full FE method.
Speaker:
Date:
Time:
Category: