Powder bed fusion (PBF) is a metal additive manufacturing process that is increasingly used in the aerospace and medical industry to build complex parts directly from computer-aided design. Due to the presence of large temperature gradients and rapid cooling rates during the processing, the PBF process is assumed to follow a rapid solidification processing route. However, the extent of deviation of the solid-liquid interface from equilibrium as a function of processing conditions has not been studied in detail for the PBF process. In this thesis, a numerical model is developed to study the interfacial characteristics as a function of processing conditions to characterize if the PBF process exhibits rapid solidification or not. The model is based on the work of Hunt et al. [1, 2, 3] and is capable of simulating cellular and dendritic growth at both low and high interface velocities. The developed model accounts for the various undercooling such as constitutional and curvature undercooling, the variation of the liquidus temperature with composition, and the partition coefficient and diffusion coefficient with temperature. Moreover, the variation of the partition coefficient and the liquidus slope with the growth velocity has also been considered in the developed model. The model is used to predict the range of primary cellular/dendritic spacing for a given set of input parameters. In addition to this, the tip undercooling, tip Péclet number and spacing Péclet numbers have also been estimated using the model to quantify the extent of deviation of the solid-liquid interface from equilibrium. A good qualitative agreement between the predicted values from the numerical model and the analytical KGT model is achieved. This new model can be used to understand the relationship between the processing conditions, material system and interfacial characteristics during the PBF process, and thus improve microstructural development during PBF processing. / Thesis / Master of Science in Materials Science and Engineering (MSMSE)

<p>The Powder Bed Fusion (PBF) process is increasingly employed by industry to fabricate complex parts with stringent standard criteria. However, fabricating parts free of defects using this process is still a major challenge. As reported in the literature, thermally induced abnormalities form the majority of generated defects and are largely attributed to thermal evolution. Various methodologies have been introduced in the literature to eliminate or mitigate such abnormalities. However, most of these methodologies are post-process in nature, lacking adaptability and customization to accommodate different geometries or materials. Consequently, they fall short of adequately addressing these challenges. Monitoring and controlling temperature, along with its distribution throughout each layer during fabrication, is an effective and efficient proxy to control the thermal evolution of the process. This, in turn, provides a real-time solution to effectively overcome such challenges. </p> <p>The objective of this dissertation is to introduce a novel online thermography and closedloop hybrid-control (NOTCH)©, an ultra-fast and practical control approach, to modify the scan strategy in metal PBF in real time. This methodology employs different mathematical-thermophysical concept-based or thermophysical-based models combined with optimization algorithms designed to optimize the printing sequence of islands/stripes/zones in order to avoid or mitigate residual stress and distortion. This methodology is adaptable to different geometries, dimensions, and materials, and is capable of being used with machines having varying ranges of specifications. </p> <p>NOTCH’s objective is to achieve a uniform temperature distribution throughout an entire layer and through the printed part (between layers) to mitigate residual stress and thermally related distortion. To attain this objective, this study explores modifying or optimizing the printing sequence of islands/stripes in an island or the strip scanning strategy. This dissertation presents three key contributions: </p> <p>First, this work introduces two potential models: the Genetic Algorithm Maximum Path (GAMP) strategy and Generalized Advanced Graph Theory. Preliminary results for a printed/simulated prototype are presented. These models, along with the Tessellation algorithm (developed in my M.Sc. thesis), were employed within NOTCH.</p> <p>Second, I developed two optimization algorithms based on the greedy and evolutionary approaches. Both algorithms are direct-derivative-free methods. The greedy optimization provides a definitive solution at each printing step, selecting the island/stripe that ensures the highest temperature uniformity. Conversely, the evolutionary algorithm seeks to obtain the final optimal solution at the end of the printing process, i.e., the printing sequence with the highest uniformity in the last printing step. This approach is inspired by the concept of Random Search algorithms, offering a non-definitive solution to find an optimal solution. </p> <p>Last, this work presents the NOTCH methodology, enabling real-time modification of printing sequences through the integration of a novel thermography methodology (developed in my M.Sc. thesis), developed models, and optimization algorithms.</p> <p><br></p>

Purdue School of Engineering and Technology, Indianapolis / Powder Bed Fusion Additive Manufacturing enables the fabrication of metal parts with complex geometry and elaborates internal features, the simplification of the assembly process, and the reduction of development time. However, the lack of consistent quality hinders its tremendous potential for widespread application in industry. This limits its ability as a viable manufacturing process particularly in the aerospace and medical industries where high quality and repeatability are critical. A variety of defects, which may be initiated during the powder-bed fusion additive manufacturing process, compromise the repeatability, precision, and resulting mechanical properties of the final part. The literature review shows that a non-uniform temperature distribution throughout fabricated layers is a significant source of the majority of thermal defects. Therefore, the work introduces an online thermography methodology to study temperature distribution, thermal evolution, and thermal specifications of the fabricated layers in powder-bed fusion process or any other thermal inherent AM process. This methodology utilizes infrared technique and segmentation image processing to extract the required data about temperature distribution and HAZs of the layer under fabrication. We conducted some primary experiments in the FDM process to leverage the thermography technique and achieve a certain insight to be able to propose a technique to generate a more uniform temperature distribution. These experiments lead to proposing an innovative chessboard scanning strategy called tessellation algorithm, which can generate more uniform temperature distribution and diminish the layer warpage consequently especially throughout the layers with either geometry that is more complex or poses relatively longer dimensions. In the next step, this work develops a new technique in ABAQUS to verify the proposed scanning strategy. This technique simulates temperature distribution throughout a layer printed by chessboard printing patterns in powder-bed fusion process in a fraction of the time taken by current methods in the literature. This technique compares the temperature distribution throughout a designed layer printed by three presented chessboard-scanning patterns, namely, rastering pattern, helical pattern, and tessellation pattern. The results confirm that the tessellation pattern generates more uniform temperature distribution compared with the other two patterns. Further research is in progress to leverage the thermography methodology to verify the simulation technique. It is also pursuing a hybrid closed-loop online monitoring and control methodology, which bases on the introduced tessellation algorithm and online thermography in this work and Artificial Neural Networking (ANN) to generate the most possible uniform temperature distribution within a safe temperature range layer-by-layer.