Efficient thermomechanical simulation of laser powder fusion 3D printing process and mechanical behavior modeling of printed structural elements
Loading...
Date
2024-07-22
Authors
Ψυχογυιός, Χαράλαμπος
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Additive Manufacturing (AM) – also known as 3D Printing – processes are a group of manufacturing processes that enable the fabrication of complex-shaped parts and are considered one of the pillars of the ongoing Fourth Industrial Revolution. Among the AM processes, Laser-based Powder Bed Fusion (L-PBF) is considered the most mature AM variation for implementation in many industrial sectors. Nevertheless, the efficient implementation of the process is hindered by many factors, mainly related to the development of rapid thermal cycles during the process and the inferior fatigue performance of the AM materials and parts. The experimental investigation of the optimum set of L-PBF process parameters consists of expensive trial-and-error procedures, while the dependence of the mechanical behavior of AM metals on the selection of process parameters is still not completely understood. Therefore, the development of reliable numerical modeling methods that predict the thermomechanical behavior and estimate the fatigue life of AM materials and parts is essential for the further implementation of L-PBF process in technological applications. However, the computational cost of these methods is enormous for practical applications. In the present study, the development of efficient methodologies for the simulation of the L-PBF process and the evaluation of the fatigue performance of the L-PBF metals/parts is presented.
To achieve the research goals, a modeling framework comprising four sub-methodologies is developed in the present thesis: the first one for small scale sub-parts simulation (meso-scale), the second for density-based optimization of the materials and parts of the L-PBF process, the third for the prediction of residual stresses/strains in larger parts without defects (part-scale) and the fourth for predicting the fatigue life of L-PBF materials and parts. More specifically, the developed thermomechanical finite element models in this thesis account for the different spatial and temporal phenomena developed during the L-PBF process. Initially, a meso-scale thermomechanical FE model aiming to model the L-PBF process of small-scale parts is developed, exploring the advantages of the adaptive mesh refinement technique – a mesh coarsening numerical technique – and it is applied for the efficient simulation of the L-PBF process. This meso-scale model predicts the temperature fields in small regions of AM parts. The thermomechanical model is demonstrated on the fabrication of the Body Centered Cubic (BCC) lattice structure unit cell. The results show that the model requires only one-quarter of the time of a more conventional simulation based on the ‘element birth and death’ technique, without any significant compromise in the accuracy of the temperature fields. Moreover, this meso-scale model can be used for the assessment of the dimensional accuracy of the structures of the L-PBF parts.
The meso-scale modeling procedures of the L-PBF process are further employed for the development of the second modeling sub-methodology for the density optimization of the L-PBF processed materials and parts. This methodology identifies the laser scan-related process parameters that lead to the minimum number of defects. This is performed by predicting melt pool characteristics for different process parameters through single laser scan thermal simulations. Then, the initial optimal parameters are used for the thermal simulation of the L-PBF fabrication of small coupons to determine their capability to produce fully dense parts. Thus, a relationship between the process parameters and the resulting porosity is determined. This approach can assist in the optimization of process parameters. The methodology is demonstrated on the optimization of processing Ti-6Al-4V and 316L alloys. The agreement between the numerical and experimental results for the dense-optimum process parameters verifies the presented approach.
The third modeling sub-methodology concerns the prediction of residual stresses and strains in large L-PBF parts. For the simulation of the L-PBF process for larger engineering components, a set of assumptions is made to avoid the explicit modeling of the laser passes and circumvent the associated time and element size constraints. The employed methodology is verified by the simulation of benchmark test cases of L-PBF components, where the applicability and limitations of the method are exposed. Guidelines for the efficient discretization of the modeled parts are also provided.
The main cause of preliminary fatigue failure of AM metals and parts is the presence of defects that act as stress raisers from which catastrophic failure initiates. As noted previously, the formation of these defects can be identified by the examination of the melt pool characteristics, which can be predicted by the thermal simulations. Thus, in the fourth sub-methodology, the detection of susceptible areas for defect formation in parts/coupons is performed by the thermal simulation of the process. The most critical defect for fatigue failure is identified and used as an input in a fracture mechanics-based model for the prediction of fatigue life estimation. The methodology is utilized for the modeling of experimental test cases of the High Cycle Fatigue (HCF) behavior of Ti-6Al-4V and 316L specimens processed by L-PBF. The predicted results show a good correlation with experimental measurements validating the presented modelling approach.
Finally, based on the examination of the results, the conclusions are outlined and recommendations for future research in the scientific area are provided.
Description
Keywords
3D printing processes, Additive manufacturing, Efficient numerical modeling and simulation, Defects, Residual stresses, Mechanical behaviour, Fatigue life prediction