Artificial intelligence (AI) is poised to significantly impact metal additive manufacturing (AM). Understanding how one might use AI in AM is challenging because AM experts are not AI experts, nor the other way around. This document introduces AI in AM and guides researchers in accessing relevant literature. It also discusses the hype surrounding AI in AM, the rush to publish peer-reviewed papers that use AI in AM, and the resulting uneven quality of the literature. Conclusions regarding the application of AI in both large and small enterprises are discussed. This document is intended to help illuminate AI in AM for * Hands-on engineers who need to quickly understand what levels of problems they might encounter when dealing with AI in AM * Engineering managers who need to stay current on emerging trends in their technical realm of responsibilities * Policymakers who may not have the relevant technical expertise * Faculty and students who want an introduction to AI in AM NOTE: SAE Edge Research Reports are intended to identify and illuminate key issues in emerging, but still unsettled, technologies of interest to the mobility industry. The goal is to stimulate discussion and work in the hope of promoting and speeding resolution of identified issues. These reports are not intended to resolve the challenges they identify or close any topic to further scrutiny.

Metal Additive Manufacturing (AM) is increasingly utilized for functional parts, often used in safety-critical applications such as jet engine components. For these applications, it is imperative that the fit, form, and function are not compromised. However, it has been shown that numerous intentional sabotage attacks are pos- sible. Understanding how sabotage attacks can be conducted is a prerequisite for their prevention and detection. This work focuses on Laser Beam Powder Bed Fusion (LB-PBF), an AM machine type dominant in the manufacturing of net-shape metal parts, and its subsystem controlling the shielding gas flow. We analyze how this essential subsystem can be manipulated to sabotage AM part performance. Our analysis shows that such sabo- tage attacks will be probabilistic, as opposed to the deterministic attacks previously discussed in the research literature. While this introduces issues with performance degradation and control over it, it is likely to also complicate the determination of intent and investigation of its root cause.

Additive manufacturing (AM), or 3D printing, of metals is transforming the fabrication of components, in part by dramatically expanding the design space, allowing optimization of shape and topology. However, although the physical processes involved in AM are similar to those of welding, a field with decades of experimental, modeling, simulation, and characterization experience, qualification of AM parts remains a challenge. The availability of exascale computational systems, particularly when combined with data-driven approaches such as machine learning, enables topology and shape optimization as well as accelerated qualification by providing process-aware, locally accurate microstructure and mechanical property models. We describe the physics components comprising the Exascale Additive Manufacturing simulation environment and report progress using highly resolved melt pool simulations to inform part-scale finite element thermomechanics simulations, drive microstructure evolution, and determine constitutive mechanical property relationships based on those microstructures using polycrystal plasticity. We report on implementation of these components for exascale computing architectures, as well as the multi-stage simulation workflow that provides a unique high-fidelity model of process–structure–property relationships for AM parts. In addition, we discuss verification and validation through collaboration with efforts such as AM-Bench, a set of benchmark test problems under development by a team led by the National Institute of Standards and Technology.

Laser powder bed fusion (L-PBF) is currently the additive manufacturing process with the widest industrial use for metal parts. Yet some hurdles persist on the way to a widespread industrial serial production, with reproducibility of the process and the resulting part properties being a major concern. As the geometry changes, so do the local boundary conditions for heat dissipation. Consequently, the use of global, geometry-independent processing parameters, which are today’s state of the art, may result in varying part properties or even defects. This paper presents a numerical simulation as a method to predict the geometry-dependent temperature evolution during the build. For demonstration, an overhang structure with varying angles towards the build platform was manufactured using Ti–6Al–4V. A calibrated infrared camera was integrated into a commercial L-PBF system to measure the temperature evolution over time for a total build height of 10 mm, and the results are used for validation of the simulation. It is shown that the simulation is capable of predicting the temperature between layers. The deviations between simulation and measurement remain in single digit range for smaller overhang structures (90°, 60° and 45°). For large overhang structures (30°), the simulation tends to over-predict the temperatures up to 15 °C. Experiments with varying process parameters showed the feasibility of energy reduction as compensation of the heat accumulation produced by overhang structures.

important dynamics in laser powder bed fusion including the identification of plasma formation mechanisms, transitions between laser-induced melting modes and stochastic precursors to defect formation. These findings were realized through implementation of electronic sensing diagnostics into an existing testbed system and harnessing capabilities developed for the time dependent analysis of laser powder bed fusion datasets. Importantly, the results advance our understanding of laser-material interactions and reveal that detection of thermionic emission can also resolve information critical to optimization of the laser powder bed fusion fabrication process. The findings are critical for advancing our fundamental understanding of the laser powder bed fusion process and can help assure the requisite fidelity of fabricated components.

During the laser-powder bed fusion ( L -PBF) process, high laser intensities, short interaction times and highly localized heat input drive large thermal gradients that result in a state of high residual stresses. Generally, the residual stresses that develop during the L -PBF process can compromise the performance of the component. Up to now, the literature has indicated that the magnitude of the residual stresses can be affected by various process parameters. In this study, all process parameters such as laser power and speed are held fixed and the focus is solely on the effect of the laser scan strategy on the three-dimensional residual stress state of L -PBF metallic components. Four Ti-6Al-4V bridge shaped components were built using island and continuous scanning patterns parallel and offset 45° from the sample axes. High-energy X-ray diffraction was used to determine the residual stress field in each of the components. Two of them were re-measured after being partially removed from the build plate. The assumptions implicit in diffraction measurements of stress are reviewed and discussed in depth because the unique microstructure associated with L-PBF Ti-6Al-4V renders the validity of those assumptions uncertain. Specifically, additional data was collected and analyzed to evaluate the relationship between grain scale and macroscopic scale stresses. The observed residual stresses were large, ½ to ¾ of the yield strength, particularly the build direction stresses near the lateral edges of the bridges. In this work, the higher stresses were observed in the bridges built via the island scan strategies, chiefly near the edges of the parts.

We present our recent development of an integrated mesoscale digital twin (DT) framework for relating processing conditions, microstructures, and mechanical responses of additively manufactured (AM) metals. In particular, focusing on the laser powder bed fusion technique, we describe how individual modeling and simulation capabilities are coupled to investigate and control AM microstructural features at multiple length and time scales. We review our prior case studies that demonstrate the integrated modeling schemes, in which high-fidelity melt pool dynamics simulations provide accurate local thermal profiles and histories to subsequent AM microstructure simulations. We also report our new mechanical response modeling results for predicted AM microstructures. In addition, we illustrate how our DT framework has been validated through modeling–experiment integration, as well as how it has been practically utilized to guide and analyze AM experiments. Finally, we share our perspectives on future directions of further development of the DT framework for more efficient, accurate predictions and wider ranges of applications.

Additive Manufacturing (AM) is a direct digital manufacturing technology increasingly used to manufacture functional parts for military and civilian applications. As AM becomes more integral to national security and economic prosperity, it also becomes an attractive cyber-physical attack target. In this paper, we focus on attacks aiming to sabotage metal parts produced using Powder Bed Fusion (PBF), a metal AM process used to manufacture near net shaped parts for safety critical systems. Specifically, we focus on the Powder Delivery System (PDS), an integral PBF subsystem. In our examination, we adopt the attacker's perspective, identify possible manipulations which can be used individually or in combination to degrade part mechanical properties. We experimentally evaluate the impact of a selected manipulation on part fatigue life. Destructive testing on two different types of stainless steel specimens, 17-4PH and 316 L, confirmed effectiveness of this attack and revealed material dependent impacts while non-destructive testing illustrated the difficulty in attack detection. Based on our analysis and experimental evaluation, we conclude that the investigated attacks have the potential to be effective against complex geometry parts such as those used in military and civilian systems while remaining undetected.

State-of-the-art metal 3D printers promise to revolutionize manufacturing, yet they have not reached optimal operational reliability. The challenge is to control complex laser-powder-melt pool interdependency (dependent upon each other) dynamics. We used high-fidelity simulations, coupled with synchrotron experiments, to capture fast multitransient dynamics at the meso-nanosecond scale and discovered new spatter-induced defect formation mechanisms that depend on the scan strategy and a competition between laser shadowing and expulsion. We derived criteria to stabilize the melt pool dynamics and minimize defects. This will help improve build reliability.

The boundary between two perfectly bonded single crystals plays a very important role in determining the deformation of the bicrystal. This work addresses the role of the grain boundary by considering the elevated hardening of a slip system due to a slip gradient. The slip gradients are associated with geometrically necessary dislocations and their effects become pronounced when a representative length scale of the deformation field is comparable to the dominant microstructural length scale of a material. A new rate-dependent crystal plasticity theory is presented and has been implemented within the finite element method framework. A planar bicrystal under uniform in-plane loading is studied using the new crystal theory. The strain is found to be continuous but non-uniform within a boundary layer around the interface. The lattice rotation is also non-uniform within the boundary layer. The width of the layer is determined by the misorientation of the grains, the hardening behavior of slip systems, and most importantly by the characteristic material length scales. The overall yield strength of the bicrystal is also obtained. A significant grain-size dependence of the yield strength, the Hall-Petch effect, is predicted.

ABSTRACT We have used a two-step (low and high temperature) strain-annealing process to evolve the grain boundary character distribution (GBCD) in fully recrystallized oxygen-free electronic (OFE) Cu bar that was forged and rolled. Orientation imaging microscopy (OIM)[1–4] has been used to characterize the GBCD after each step in the processing. The fraction of special grain boundaries, “special fraction,” was ∼70% in the starting recrystallized material. Three different processing conditions were employed: high, moderate, and low temperature. The high-temperature process resulted in a reduction in the fraction of special grain boundaries while both of the lower temperature processes resulted in an increase in special fraction up to 85%. Further, the lower temperature processes resulted in average deviation angles from exact misorientation, for special boundaries, that were significantly smaller than observed from the high temperature process. Results indicate the importance of the low temperature part of the two-step strain-annealing process in preparing the microstructure for the higher temperature anneal and commensurate increase in the special fraction.