Additive Manufacturing for Energy Applications II: Modelling
Sponsored by: TMS Structural Materials Division, TMS: Nuclear Materials Committee
Program Organizers: Isabella Van Rooyen, Pacific Northwest National Laboratory; Subhashish Meher, Pacific Northwest National Laboratory; Indrajit Charit, University of Idaho; Michael Kirka, Oak Ridge National Laboratory
Tuesday 2:00 PM
February 25, 2020
Room: 9
Location: San Diego Convention Ctr
Session Chair: Indrajit Charit, University of Idaho; Wen Jiang, Idaho National Laboratory
2:00 PM Invited
Progress in Additive Manufacturing in South Africa: Sisa Pityana1; 1CSIR Main Campus
Additive Manufacturing (AM) has come to change the way products are designed and manufactured due to the flexibility it offered the designer. In the past, products were designed through proper consideration of ease of manufacturing the product. This gives the designer, the freedom of design with the sole concentration on the work ability of the product. The CSIR in South Africa has developed the worlds’ largest powder bed system, which is being tested and commissioning. The machine can produce parts as big as 2 m x 0.6 m x 0.6 m using a 5 kW laser. This talk will focus on the laser additively manufactured parts using mainly Ti6Al4V, paying attention on mechanical and metallurgical integrity of the components. Simulations and experimental work will be compared. This is to generate the needed knowledge on which manufacturing process to select at any particular point in time and for what application.
2:30 PM
Multi-physics Simulation to Model Melt Pool in Directed Energy Deposition Process for Nuclear Fuels: Wen Jiang1; Jeong-Hoon Song2; Isabella Rooyen1; 1Idaho National Laboratory; 2University of Colorado Boulder
There is significant interest in applying additive manufacturing (AM) techniques to manufacture nuclear fuel systems. Compared to traditional fuel production techniques, AM technique greatly simplifies the fabrication procedure, reduces the production time and cost, and provides additional flexibility to nuclear fuel design for better thermal and mechanical performance. Modeling and simulation can provide the insight for a rational design by addressing the basic science and engineering needs for AM process, leading to a mechanistic understanding of processing-structure-property relationships, which are fundamental to design and develop innovative fabrication techniques for nuclear fuels. At mesoscale, a high-fidelity numerical model including heat transfer, fluid flow and mass transport, is developed in Multiphysics Object Oriented Simulation Environment (MOOSE) to simulate the melt pool physics. The temperature, chemical composition and solidification rate of melt pool can be predicted and thereafter used in lower-length scale model to simulate microstructure evolution.
2:50 PM
Mechanical Response and Reduced-order Simulations of Additively Manufactured Metallic Lattice Structures: Connie Dong1; Sara Messina1; Matthew Begley1; 1University of California, Santa Barbara
While additive manufacturing provides an exciting pathway to novel lattice-based components, difficulties arise in predicting their performance due to process-structure-property coupling arising from printing, resulting in extended design cycle times which limits full adoption of these materials to market. Experiments designed to establish structure-property relationships in EBM Ti-6Al-4V lattice primitives of struts, multi-strut intersections (“nodes”), and multi-node “cells” will be presented. These experiments illustrate that the resultant performance of these primitives is strongly dependent on their orientation relative to the print direction due to orientation-dependent defects and geometry. Based on these experiments, a reduced-order simulation approach that adapts conventional beam elements to account for the spread of plasticity near defects and in strut intersections has been developed. These results are demonstrated to be in excellent agreement with conventional full-dimensional FEA and are orders of magnitude faster.
3:10 PM
Active Force Regimes in Powder Spreading: Nicholas Cunningham1; Noah Philips1; Yifei Ma2; T. Matthew Evans2; 1Ati Specialty Alloys And Components; 2Oregon State University
Defect-free powder bed additive manufacturing requires well behaved powder spreading to achieve uniform, thin layers. Rotary drum testing on powder feedstock can quantify the effect of particle size distribution, absolute moisture content, and powder morphology on powder flow, but there is a desire to fundamentally understand how these powder characteristics dictate spreading quality. In this study, a discrete element method (DEM) model captures the cohesive mechanical, electrostatic, and capillary forces present in powder bed systems. Simple experiments with refractory powders are used to calibrate the model, while rotary drum and spreading tests are used for model validation.
3:30 PM Break
3:50 PM
Digital Twins of Additive Manufacturing Processes for the Optimization and Control of Builds: Dayalan Gunasegaram1; Anthony Murphy1; Patrick O'Toole1; Daniel East1; 1CSIRO
In the end-to-end digitalisation of manufacturing foreseen in an Industry 4.0 economy, machines will be able to act autonomously based on insights developed by artificial intelligence and advanced analytics using a wealth of real-time process data gathered by sensors. For robust decision-making, however, process intelligence that is sufficiently in-depth is required. Such intelligence may be provided by physics-based computational models, which are called ‘digital twins’ (DT) of the processes. Advance knowledge of how a process would unfold will equip the machine to ensure the expensive part is made right first time and every time. This is achieved by using the optimum process parameters suggested by the DT and checking the progress of the build against the optimum path and correcting if it is outside tolerance levels. In this talk, hurdles to the development of the DTs for additive manufacturing processes will be discussed, along with potential solutions to these issues.
4:10 PM
Numerical Simulation and High-speed Photography Characterization of Powder Delivery During LENS® Additive Manufacturing for Metal Matrix Composites: Sen Jiang1; Baolong Zheng1; James Haley1; Bingqing Chen2; Jiayu Liang2; Shuai Huang2; Enrique Lavernia1; Julie Schoenung1; 1University of California, Irvine; 2Beijing Institute of Aeronautical Materials
Inconel 718 and Inconel 718+TiC MMC were fabricated using Laser Engineered Net Shaping (LENS®). The size and morphology of the feed stock powder, carrier gas flow condition and nozzle geometry can all affect powder delivery behavior. In this work, high-speed photography with 100k frames per second was utilized to observe the particle flight from the nozzle exit to the substrate and interaction with the molten pool. Particle position, flight trajectory and density distribution for both Inconel 718 and TiC powder were obtained. The captured videos revealed distinct particle cloud patterns and particle flow behavior for Inconel 718 and TiC, due to the differences in particle size and morphology. A three-dimensional COMSOL Multiphysics® model was constructed to investigate the gas flow field, particle-gas interactions, particle-particle interactions and particle-wall collisions during LENS® deposition. The calculated particle trajectories and density distributions were validated using the captured high-speed video.
4:30 PM
Design of Efficient Additive-manufactured Heat Sinks via Conjugate Heat Transfer Modeling and Topology Optimization: Basil Paudel1; Mohammad Masoomi2; Scott Thompson3; 1Auburn University; 2ANSYS Inc.; 3Kansas State University
The higher design flexibility offered by additive manufacturing (AM) allows for radical improvements in the design and functionality of legacy parts for several energy applications. In this current study, a cross-flow/air-cooled heat sink is designed and optimized using the ANSYS® topology optimization module. Unlike many other design/optimization approaches for AM, the current method employs a multi-objective scheme for mass reduction and conjugate heat transfer maximization. A gradient-based adjoint solver is used along with ANSYS’s computational fluid dynamics (CFD) package to seek the optimal shape for a given set of heat sink operating conditions. Pin-fin array and gyroid-based heat sink concepts were used for ‘seeding’ the multi-objective topology optimization routines and the results are compared and discussed. Topology design and operating (boundary condition) variables were varied to elucidate major design sensitivities. In all cases, the optimized heat sink design performed significantly better than its benchmark in terms of thermal efficiency.