Residual stress modeling of thermally-driven welding and additive manufacturing
- Eigenspannungsmodellierung von thermisch angetriebenem Schweißen und additiver Fertigung
Ali, Baharin Rahim Ali; Markert, Bernd (Thesis advisor); Kiefer, Björn (Thesis advisor)
Aachen : RWTH University Aachen (2023)
Dissertation / PhD Thesis
In: Report. IAM, Institute of General Mechanics 20
Page(s)/Article-Nr.: 1 Online-RessourceReport number: IAM-20
Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2023
Abstract
In the industrial applications, thermally-driven manufacturing approaches such as welding and additive manufacturing are widely used to assemble and create metallic parts. The Gas tungsten arc welding (GTAW) is fusion welding process used to join two or more metallic components together. On the other hand, additive manufacturing (AM) is a revolutionary manufacturing technique used to construct objects of sophisticated geometries by successively melting layers of metallic powder. In both GTAW and AM, the material being treated undergoes complex multiphysical phenomena driven by the rapid heating-cooling cycles. Therefore, understanding and investigating these fusion processes are essential to gain insight into the coupled multiphysical phenomena occurring during melting, remelting, and solidification processes. One of the main aims in the underlying work is to establish an approach to understanding the distribution mechanism of stresses and deformations. It has been reported that modeling the thermomechanical responses in welding and AM is very time consuming and computationally demanding. Therefore, simplifications are required to improve the performance and allow the modeling of larger domains with a sufficient accuracy. In this thesis, a thermomechanical modeling approach is developed to understand the fundamental physics of thermally-driven processes with the aim of predicting the temperature field and the final state of residual stresses in GTAW and AM. To this end, a macroscopic material modeling approach is developed based on a continuum mechanical framework in which the welded or AM part is represented as a homogenized continuous medium. Following the thermodynamically consistent formulation, the approach leads to volume-coupled nonlinear partial differential equations (PDEs). The formulations employ the phase-field method (PFM) in a novel way to describe the state of the material and to track its phase change from solid (hard) and unconsolidated powder to liquid (soft) and finally to a solidified state. Furthermore, PFM enables the definition of the material properties as functions of the material state, i.e., hard, soft, or powder state. The material states (phases) are separated by a diffuse interface that allows smooth transition of these states across the interface. In particular, the phase-field and phase-field history variables are utilized to propose expressions for phasefield-dependent material properties. In this treatment, a realistic description of the material parameters is achieved based on the material state, which leads to a reduction in the modeling complexity. The formulation of the phase-field thermo-elastoplastic model leads to the coupled governing PDEs, which are complemented by constitutive relations and evolution equations. Regarding the elastoplastic response in metals, two different plasticity models are described and implemented, i.e., the return mapping method and the incremental method, following the J2 theory with the isotropic strain hardening law. Furthermore, numerical investigations are performed to compare the performance of the implemented models using the open-access finite element package FEniCS Project. To validate our modeling approaches, representative examples of two and three-dimensional GTAW and selective laser melting are implemented and solved using the finite element method. In these initial-boundary-value problems, the temperature fields and residual stresses are compared with reference solutions from the literature.
Identifier
- DOI: 10.18154/RWTH-2023-04243
- RWTH PUBLICATIONS: RWTH-2023-04243