The objective of this work is to identify failure modes and detect the onset of process anomalies in additive manufacturing (AM) processes, specifically focusing on fused filament fabrication (FFF). We accomplish this objective using advanced Bayesian nonparametric analysis of in situ heterogeneous sensor data. Experiments are conducted on a desktop FFF machine instrumented with a heterogeneous sensor array including thermocouples, accelerometers, an infrared (IR) temperature sensor, and a real-time miniature video borescope. FFF process failures are detected online using the nonparametric Bayesian Dirichlet process (DP) mixture model and evidence theory (ET) based on the experimentally acquired sensor data. This sensor data-driven defect detection approach facilitates real-time identification and correction of FFF process drifts with an accuracy and precision approaching 85% (average F-score). In comparison, the F-score from existing approaches, such as probabilistic neural networks (PNN), naïve Bayesian clustering, support vector machines (SVM), and quadratic discriminant analysis (QDA), was in the range of 55–75%.

The goal of this work is to achieve the defect-free production of parts made using Additive Manufacturing (AM) processes. As a step towards this goal, the objective is to detect flaws in AM parts during the process by combining predictions from a physical model (simulation) with in-situ sensor signatures in a machine learning framework. We hypothesize that flaws in AM parts are detected with significantly higher statistical fidelity (F-score) when both in-situ sensor data and theoretical predictions are pooled together in a machine learning model, compared to an approach that is based exclusively on machine learning of sensor data (black-box model) or physics-based predictions (white-box model). We test the hypothesized efficacy of such a gray-box model or digital twin approach in the context of the laser powder bed fusion (LPBF) and directed energy deposition (DED) AM processes. For example, in the DED process, we first predicted the instantaneous spatiotemporal distribution of temperature in a thin-wall titanium alloy part using a computational heat transfer model based on graph theory. Subsequently, we combined the preceding physics-derived thermal trends with in-situ temperature measurements obtained from a pyrometer in a readily implemented supervised machine learning framework (support vector machine). We demonstrate that the integration of temperature predictions from an ab initio heat transfer model and in-situ sensor data is capable of detecting flaws in the DED-produced thin-wall part with F-score approaching 90%. By contrast, the F-score decreases to nearly 80% when either temperature measurements from the in-situ sensor or temperature distribution predictions from the theoretical model are used alone by themselves. This work thus demonstrates an early foray into the digital twin paradigm for real-time process monitoring in AM via seamless integration of physics-based modeling (simulation), in-situ sensing, and data analytics (machine learning).

Hybrid additive manufacturing (hybrid-AM) has described hybrid processes and machines as well as multimaterial, multistructural, and multifunctional printing. The capabilities afforded by hybrid-AM are rewriting the design rules for materials and adding a new dimension in the design for additive manufacturing (AM) paradigm. This work primarily focuses on defining hybrid-AM in relation to hybrid manufacturing (HM) and classifying hybrid-AM processes. Hybrid-AM machines, materials, structures, and function are also discussed. Hybrid-AM processes are defined as the use of AM with one or more secondary processes or energy sources that are fully coupled and synergistically affect part quality, functionality, and/or process performance. Historically, defining HM processes centered on process improvement rather than improvements to part quality or performance; however, the primary goal for the majority of hybrid-AM processes is to improve part quality and part performance rather than improve processing. Hybrid-AM processes are typically a cyclic process chain and are distinguished from postprocessing operations that do not meet the fully coupled criterion. Secondary processes and energy sources include subtractive and transformative manufacturing technologies, such as machining, remelting, peening, rolling, and friction stir processing (FSP). As interest in hybrid-AM grows, new economic and sustainability tools are needed as well as sensing technologies that better facilitate hybrid processing. Hybrid-AM has ushered in the next evolutionary step in AM and has the potential to profoundly change the way goods are manufactured.

Although complex geometries are attainable with additive manufacturing (AM), a major barrier preventing its use in mission-critical applications is the lack of geometric accuracy of AM parts. Existing geometric dimensioning and tolerancing (GD&T) characteristics are defined based on simple landmark features, and thus, need to be customized to capture the subtle difference in parts with complex geometries. Hence, the objective of this work is to quantify the geometric deviations of additively manufactured parts from a large data set of laser-scanned coordinates using an unsupervised machine learning (ML) approach called the self-organizing map (SOM). The central hypothesis is that clusters recognized by the SOM correspond to specific types of geometric deviations, which in turn are linked to certain AM process conditions. This hypothesis is tested on parts made while varying process conditions in the fused filament fabrication (FFF) AM process. The outcomes of this research are as follows: (1) visualizing and quantifying the link between process conditions and geometric accuracy in FFF and (2) significantly reducing the amount of point cloud data required for characterizing of geometric accuracy. The significance of this research is that this unsupervised ML approach resulted in less than 3% of over 1 million data points being required to fully quantify the part geometric accuracy.

The goal of this work is to understand the effect of process conditions on lack of fusion porosity in parts made using laser powder bed fusion (LPBF) additive manufacturing (AM) process, and subsequently, to detect the onset of process conditions that lead to lack of fusion-related porosity from in-process sensor data. In pursuit of this goal, the objectives of this work are twofold: (1) quantify the count (number), size and location of pores as a function of three LPBF process parameters, namely, the hatch spacing (H), laser velocity (V), and laser power (P); and (2) monitor and identify process conditions that are liable to cause porosity through analysis of in-process layer-by-layer optical images of the build invoking multifractal and spectral graph theoretic features. These objectives are important because porosity has a significant impact on the functional integrity of LPBF parts, such as fatigue life. Furthermore, linking process conditions to defects via sensor signatures is the first step toward in-process quality assurance in LPBF. To achieve the first objective, titanium alloy (Ti–6Al–4V) test cylinders of 10 mm diameter × 25 mm height were built under differing H, V, and P settings on a commercial LPBF machine (EOS M280). The effect of these process parameters on count, size, and location of pores was quantified based on X-ray computed tomography (XCT) images. To achieve the second objective, layerwise optical images of the powder bed were acquired as the parts were being built. Spectral graph theoretic and multifractal features were extracted from the layer-by-layer images for each test part. Subsequently, these features were linked to the process parameters using machine learning approaches. Through these image-based features, process conditions under which the parts were built were identified with the statistical fidelity over 80% (F-score).

The objective of this work is to develop and apply a spectral graph theoretic approach for differentiating between (classifying) additive manufactured (AM) parts contingent on the severity of their dimensional variation from laser-scanned coordinate measurements (3D point cloud). The novelty of the approach is in invoking spectral graph Laplacian eigenvalues as an extracted feature from the laser-scanned 3D point cloud data in conjunction with various machine learning techniques. The outcome is a new method that classifies the dimensional variation of an AM part by sampling less than 5% of the 2 million 3D point cloud data acquired (per part). This is a practically important result, because it reduces the measurement burden for postprocess quality assurance in AM—parts can be laser-scanned and their dimensional variation quickly assessed on the shop floor. To realize the research objective, the procedure is as follows. Test parts are made using the fused filament fabrication (FFF) polymer AM process. The FFF process conditions are varied per a phased design of experiments plan to produce parts with distinctive dimensional variations. Subsequently, each test part is laser scanned and 3D point cloud data are acquired. To classify the dimensional variation among parts, Laplacian eigenvalues are extracted from the 3D point cloud data and used as features within different machine learning approaches. Six machine learning approaches are juxtaposed: sparse representation, k-nearest neighbors, neural network, naïve Bayes, support vector machine, and decision tree. Of these, the sparse representation technique provides the highest classification accuracy (F-score > 97%).