Aurrerapenak hari bidezko fabrikazio gehigarriko ereduetan

Abstract

This work analyzes the modeling of wire-based additive manufacturing processes (WAAM and DED) from a multiscale perspective, considering the interaction between the micro, meso, and macroscales to achieve a comprehensive understanding of the process. Although the relevance of all scales is acknowledged, the core focus of the work is placed on the mesoscale, as bead cross-sectional geometry, meltpool behavior, and local temperature distribution largely govern the final process outcomes. Several modeling approaches used at the mesoscale are examined, including analytical formulations, thermo-capillary and gravitational balance models, physics-based computational fluid dynamic (CFD) simulations, and recent data-driven machine learning (ML) and hybrid approaches. The analysis shows that CFD models are the most suitable in academic contexts due to their ability to capture detailed and coherent physics, despite their high computational cost; conversely, ML-based approaches offer greater computational efficiency and practical applicability in industrial environments. From the microscale perspective, the role of solidification phenomena is outlined, while at the macroscale, the influence of heat propagation and distortion evolution is highlighted. Ultimately, it is concluded that achieving an integral understanding of the process requires consistent data transfer and compatibility across the three scales.