This article presents labeled data collected from video footage of additive manufacturing processes. This computer vision dataset contains 90 labeled additive manufacturing video segments across four distinct additive manufacturing processes, amounting to 900 labeled, object instance video frames. The technologies covered by the dataset include laser hot-wire additive manufacturing (LHW-DED), plasma arc welding (PAW), tungsten inert gas wire arc additive manufacturing (TIG-WAAM), and polymer-based additive manufacturing (visPolymer and irPolymer). For each of the manufacturing processes, build deposition video was captured and then randomly sampled into 10 frame-long video segments. Depending on the manufacturing process, two of the following four object instance classes are labeled in each frame: Melt Pool, Feed Wire, Nozzle, and Material. Since the dataset’s directory organization follows that of common video object segmentation (VOS) datasets, such as coMplex video Object SEgmentation (MOSE), MOSEv2, Densely Annotated Video Segmentation (DAVIS), and YouTube-VOS, it can be easily integrated into other VOS model training pipelines for foundation model fine-tuning, training, or inference capability testing. This dataset provides researchers access to labeled additive manufacturing object instances for VOS tasks.

Directed energy deposition (DED) is an additive manufacturing process that is being rapidly adopted by industry and is well suited for the fabrication of complex components in a variety of metal alloys. In laser cladding systems such as DED, powder is blown in a stream to a metal substrate coincident with a laser necessary to deposit molten metal with 3D spatial control. The focus of both the laser and the powder stream are crucial, and best deposition occurs at a predetermined standoff height between the build surface and the print head. Generally, no monitoring of this distance is implemented in commercial DED systems. Due to potential over or under building, the standoff height often changes over time but tends to self-correct. However, inexpensive and minimally intrusive methods to identify optimal standoff are required to provide real-time control to maintain the optimal distance. The present work explores the quantification of the focus of the three-color channels of a coaxial camera to determine the standoff height. An experiment was performed in which a 254 mm wall is built and the standoff height, initially 5.0 mm below the optimal position, was then intentionally increased every 25.4 mm of wall length by an amount of 1.0 mm to a final position 7.0 mm above optimal. Computer vision is demonstrated to monitor the amount of focus in each color band and estimate standoff distance. A response can be calculated in under 40 ms using simple hardware and can work in most laser-based DED systems.

Hybrid manufacturing, or the combination of additive and subtractive manufacturing within a single build volume is transforming the way products are being fabricated. Additive manufacturing confers unprecedented freedom of design and reduced material usage while enabling serial customization. Subtractive manufacturing provides superior surface finish and improved dimensional accuracies. Interwoven, these two digital manufacturing paradigms are enabling the rapid manufacturing of complex, highly accurate, and customized geometries in a diversity of high-performance alloys. In situ monitoring, heavily relied upon in either additive or subtractive, becomes even more crucial with the interplay of the two processes in a single combined build sequence. Moreover, challenges that do not exist in either process alone can now have a dramatic impact on final part quality: (1) a large amount of heat is generated during additive manufacturing deposition, which is primarily dissipated into the machine tooling and impacts accuracy by deforming the material and causing misaligned machining; (2) microstructure, mechanical properties, and residual stresses are the result of the complex thermal histories generated by additive manufacturing and can impact subsequent cutting performance. This comprehensive review considers previous research in monitoring and providing closed-loop control in both additive and subtractive manufacturing separately and then considers the implications of the effectiveness of these monitoring techniques when additive and subtractive processes are integrated together.

Neutron diffraction is a useful technique for mapping residual strains in dense metal objects. The technique works by placing an object in the path of a neutron beam, measuring the diffracted signals and inferring the local lattice strain values from the measurement. In order to map the strains across the entire object, the object is stepped one position at a time in the path of the neutron beam, typically in raster order, and at each position a strain value is estimated. Typical dwell times at neutron diffraction instruments result in an overall measurement that can take several hours to map an object that is several tens of centimeters in each dimension at a resolution of a few millimeters, during which the end users do not have an estimate of the global strain features and are at risk of incomplete information in case of instruments outages. In this paper, we propose an object adaptive sampling strategy to measure the significant points first. We start with a small initial uniform set of measurement points across the object to be mapped, compute the strain in those positions and use a machine learning technique to predict the next position to measure in the object. Specifically, we use a Bayesian optimization based on a Gaussian process regression method to infer the underlying strain field from a sparse set of measurements and predict the next most informative positions to measure based on estimates of the mean and variance in the strain fields estimated from the previously measured points. We demonstrate our real-time measure-infer-predict workflow on additively manufactured steel parts—demonstrating that we can get an accurate strain estimate even with 30%–40% of the typical number of measurements—leading the path to faster strain mapping with useful real-time feedback. We emphasize that the proposed method is general and can be used for fast mapping of other material properties such as phase fractions from time-consuming point-wise neutron measurements.

Directed Energy Deposition (DED), a class of additive manufacturing techniques, has seen rapid growth over the last decade for potential applications in aerospace, medical devices, and energy systems. Despite notable progress in the research and development of AM, control and mitigation of residual stress during DED remains a challenge. In this work, we propose a novel approach that can be used for the mitigation of residual stresses in additively manufactured components. Specifically, we propose to mitigate the residual stress state of as-deposited components using alloy design, engineering of solid-state transformations, and the introduction of both hard and soft metallic phases. We demonstrate this strategy with a model system consisting of pure Fe and Fe–Cu. Experimental results indicate that residual stresses can be successfully manipulated by adjusting the alloy composition as a soft metallic phase can accommodate plastic deformation. Moreover, our findings suggest that the solid-state transformations experienced by the Fe and Fe-rich phases contribute to the observed differences in magnitude and location of residual stresses. This study is the first to suggest using residual stress as an engineering criterion in the design of alloys for metal additive manufacturing.

The Advanced Materials and Manufacturing Technologies (AMMT) Program is aimed at developing cross-cutting technologies in support of a broad range of nuclear reactor parts, and to maintain U.S. leadership in materials and manufacturing technologies for nuclear energy applications. The overarching vision of the AMMT program is to accelerate the development, qualification, demonstration and deployment of advanced materials and manufacturing technologies to enable reliable and economical nuclear energy. One of its three goals is to target big challenges and game-changing technologies, to realize the mission and vision of AMMT program.Based on this context, this multi-year work package focuses on understanding the current state of large-scale additive manufacturing (AM) technology for the deposition of 316L stainless steel materials for final components and mild steel for use in nuclear manufacturing processes. The targeted AM modality is directed energy deposition (DED), capable of fabricating components on the size scale of meters including valves, pumps, impellers. etc. that are challenging or difficult to source, especially when developing new systems or replacing obsolete components. Accordingly, the current writeup aims at providing a baseline literature survey on structure-property relationships in mild steel and 316L alloys. Also, information on preliminary trials to date involving these two alloys show tremendous potential of printing parts having complex geometry and thin- walled structures, such as nuclear valve and Hot isostatic Press (HIP) can, using wire based (Wire Arc Additive Manufacturing and Hybrid Additive Manufacturing) as well as blown powder DED machines. All of this is aimed towards (i) demonstrating the ability to fabricate large components for pressure boundary applications relevant to the nuclear community and nuclear manufacturing technology, and (ii) understanding the effect of different manufacturing technology on AM can production and post HIPed material for nuclear applications.Another target of this writeup is to compile various in-situ monitoring tools that have been incorporated for different DED AM modalities, in order to understand process variability during the entire fabrication process. This can be correlated with processing-structure-property response surfaces and would add confidence around process quality verification and ultimately component certification for nuclear applications.

Residual stresses affect the performance and reliability of most manufactured goods and are prevalent in casting, welding, and additive manufacturing (AM, 3D printing). Residual stresses are associated with plastic strain gradients accrued due to transient thermal stress. Complex thermal conditions in AM produce similarly complex residual stress patterns. However, measuring real-time effects of processing on stress evolution is not possible with conventional techniques. Here we use operando neutron diffraction to characterize transient phase transformations and lattice strain evolution during AM of a low-temperature transformation steel. Combining diffraction, infrared and simulation data reveals that elastic and plastic strain distributions are controlled by motion of the face-centered cubic and body-centered cubic phase boundary. Our results provide a new pathway to design residual stress states and property distributions within additively manufactured components. These findings will enable control of residual stress distributions for advantages such as improved fatigue life or resistance to stress-corrosion cracking.

Computer aided manufacturing (CAM) techniques for directed energy deposition (DED) affect the material properties of the manufactured component based on the scan strategy used. This research investigates the material characteristics of turning-style toolpath strategies to generate axisymmetric components with additive manufacturing (AM), referred to in this research as additive turning. This novel approach leverages existing CAM technology for turning, where the component rotates around a stationary cutting tool, to generate toolpath trajectories for DED with varying wall-thicknesses and controlled deposition angles. This strategy allows for entire components to be deposited in one continuous deposition, resulting in reduced cycle-time and improved material usage efficiency compared to conventional AM strategies where the beam is switched off at the end of every layer. Results from this study show that the use of additive turning can produce over 99 % dense components with less variation and anisotropy in texture and hardness, as well as a lower variation in elongation to failure when compared to conventional strategies. This research highlights that various CAM strategies could be deployed for AM to improve process efficiency or enable localized control over part performance.

As in welding, directed energy deposition (DED) additive manufacturing (AM) generates complex residual stresses and distortions commensurate with the complexity of the scan pattern used for deposition. To date, measuring DED distortions on complex geometries has only been achieved post process, discarding the complex thermomechanical history that leads to that final material state. In this work, surround stereo digital image correlation (DIC) is used to 3D map surfaces and strain tensors in-situ in a powder-blown laser DED system. Infrared thermography is then projected onto these surfaces to record the full thermomechanical history of printed parts. DIC presents a unique challenge to DED AM, as no part exists at the beginning of deposition, which (a) prevents application of an appropriate speckle pattern and (b) denies the user a zero strain reference frame. Solutions to these problems are proposed and their limitations explored herein. In sum, this work presents a relatively low-cost solution to monitoring and optimizing the unique temporal artifacts induced by complex scan strategies that was previously unobtainable.

The Transformational Challenge Reactor program is leveraging additive manufacturing technologies to fabricate the nuclear components required to assemble a microreactor core. Compared with traditional manufacturing processes, additive manufacturing allows for direct observation of the interior of the component during manufacturing. This unique capability promises significant possibilities for creating a new paradigm for nuclear component qualification by leveraging in-situ process data. This report describes FY21 efforts to predict material tensile properties based on data collected during the laser powder bed fusion printing process. The primary focus of this report is the test campaign designed to generate the large quantities of training data required to implement artificial intelligence algorithms that can predict these material properties. Preliminary prediction results and a demonstration of the overall data collection, analysis, and visualization pipeline are also provided.

Machine Learning (ML) and Digital Twins (DT) are at the heart of today’s different industries, ranging from advanced manufacturing to biomedical systems to resilient ecosystems, civil infrastructures, smart cities, and healthcare. They have become indispensable for solving complex problems in science, engineering, and technology development. The purpose of the MMLDT-CSET 2021 conference is to facilitate the transition of ML and DT from fundamental research to mainstream fields and technologies through advanced data science, mechanistic methods, and computational technologies. This 3-day conference features technical tracks of emerging ML-DT fields and applications, special public lectures, short courses, and demonstrations. The conference will be held in a hybrid format, featuring both on-site and virtual sessions.

The Transformational Challenge Reactor (TCR) program is leveraging additive manufacturing (AM) technologies to fabricate nuclear components which will be assembled into a fully functional microreactor core. Compared with traditional manufacturing technologies, AM technologies allow (1) real-time observation of the manufacturing process at a much higher resolution using in-situ monitoring technologies to capture the sensor signatures that scientifically describe each event occurring over time and space and (2) validation of the manufacturing process quality using domain-informed data analytics techniques as a potential qualification and certification methodology for the final component. This report provides an update on the program work on in-situ and ex-situ data correlation and associated data analytics results. Examples are provided to illustrate progress with respect to laser powder bed fusion (L-PBF), binder jetting, computed tomography (CT) reconstruction, and mechanical testing. Elements of the Digital Thread and data management infrastructure are discussed in the main document, and an extensive supplemental appendix is provided detailing the Digital Platform, as well as its implementation and subcomponents. In conclusion, the path forward for the next fiscal year is also discussed.

The Transformational Challenge Reactor (TCR) program is leveraging additive manufacturing (AM) technologies to fabricate nuclear components to be assembled into a fully functional microreactor core. Compared with traditional manufacturing technologies, AM technologies allow (1) observation of the manufacturing process at a much higher resolution in real-time using in situ monitoring technologies to capture the sensor signature that scientifically describes each event occurring over time and space and (2) validation of the manufacturing process quality using domain-informed data analytics techniques as a potential qualification and certification methodology for the final component. This report provides an update on the program work on binder jetting in situ process monitoring and associated data analytics results, as well as sample placement and tracking for the subsequent chemical vapor infiltration (CVI) process. Examples are provided to illustrate the progress. Elements of the Digital Thread and data management are discussed in the main document, and an extensive supplemental material section is provided detailing the Digital Platform, as well as its implementation and components. In conclusion the path forward for the next fiscal year is discussed.

program work on laser powder bed fusion in-situ process monitoring and associated data analytics results. Examples are provided to illustrate the progress. Elements of the Digital Thread and data management are discussed in the main document, and an extensive supplemental material section is provided detailing the Digital Platform, as well as its implementation and components. In conclusion the path forward for the next fiscal year is discussed.

In powder based Laser-Directed Energy Deposition (L-DED), an incident laser melts a millimeter scale pool of metal, into which feedstock powder is sprayed. Previous high speed video reveals that powders are trapped by surface tension and float for a brief residence time before melting, directly contributing to surface roughness and loss of mass capture efficiency. In this work, influencing factors on this behavior are investigated with numerical models through coupling a three phase (gas, liquid, solid) Computational Fluid Dynamics (CFD) model with applied surface tension to a heat transfer model and observing the melting dynamics of an individual powder particle of stainless steel 316 L. Sensitivity of residence time to particle size, impact velocity, melt pool and particle temperature, surface tension, and material thermophysical properties are investigated. It is found that simulations can be condensed into a simplified analytic equation, providing a rapid, explicit estimation of residence time. The demonstrated sensitivity of L-DED to powder scale surface wettability phenomena highlights a fundamental mechanistic reason why control of feedstock powder properties is essential for reliable system behavior.

Author(s): Haley, James Cameron | Advisor(s): Lavernia, Enrique J | Abstract: Laser Directed Energy Deposition (L-DED) Additive Manufacturing (AM) offers unprecedented flexibility in direct fabrication of metallic components in a way that can be readily integrated with existing CNC subtractive machining technologies. The core building block of the technology is the melt pool, the dynamic bead of molten material established by the energy equilibrium between incident laser energy and thermal dissipation. While the unique solidification microstructure of the melt pool has attracted intense scrutiny, the mechanisms determining how mass is originally incorporated into the melt pool have been less well studied. In this work, three new tools are applied to the task of broadening the understanding mass capture behaviors. First, over long time scales it was observed that mass capture efficiency evolves over the course of depositing many layers as machine conditions change; a non-empirical model constructed to track this revealed self-stabilizing behavior in working distance in open-loop control systems. Second, on very short time scales, high speed videography was employed to understand what happens at the moment of impact between a feedstock powder particle and the melt pool. It was revealed that particles are captured by surface tension before fully melting. Third, this particle retention time was investigated with numeric simulation to highlight its relationship to particle size, impact velocity, thermal distributions and wettability.