Gregory J. Wagner

The extent to which a three-dimensional extrusion of a two-dimensional microstructure (2DM) can reproduce the thermomechanical response of a fully three-dimensional microstructure (3DM) is investigated. A multi-physics numerical framework coupling computational thermal fluid dynamics (CFD), phase-field (PF) solidification, and thermo-elasto-viscoplastic finite element (TEVP-FE) modeling is employed to simulate a single laser line scan on a 316L stainless steel substrate. The temperature evolution obtained from CFD and the final microstructure predicted by PF simulations are used to perform two TEVP-FE simulations that differ only in the representation of the microstructure: 2DM and 3DM. Residual stresses, plastic strains, and Nye's tensor are compared at both local and statistical levels. The 2DM approximation captures the overall spatial evolution trends and the order of magnitude of residual stresses, but it does not reproduce the localization of shear stresses, plastic strains, and Nye's tensor, which is strongly influenced by the 3D grain morphology. Nevertheless, comparison of grain surface-averaged quantities on the lasered surface shows that the intergranular mechanical fields predicted by 2DM and 3DM match well in magnitudes and evolution trends. These results quantify the advantages and limitations of 2D microstructure approximations and provide guidance on the model complexity required for predicting intergranular mechanical fields at local and statistical levels.

Metal additive manufacturing (AM) offers a unique opportunity for production of advanced materials and complex geometries. However, variability in microstructure and properties challenges conventional approaches to design, process optimization, qualification, and materials selection. Modeling and simulation can improve understanding of AM processing and materials, but also poses major challenges for existing computational methods. Simultaneously, modern scientific computing hardware has become increasingly complex, most notably with the adoption of hybrid architectures such as Graphical Processing Units (GPUs). If appropriately utilized, emerging computational capabilities provide an opportunity to reveal new insight into AM processing and the resulting material structure and properties. In this review we describe the computational AM landscape, identify critical gaps, and highlight opportunities to impact the development and application of AM. First, the requirements and challenges of representative AM problem statements will be defined. These problems range from scientific studies to industrial applications and are designed to capture the breadth of challenges facing the AM community. Next, the current state of AM modeling and simulation is evaluated, broken down by enabling hardware and software, process simulation, microstructure simulation, and property simulation. Each section describes the diversity of simulation approaches and associated trade-offs in physical fidelity and computational expense. Each area is then assessed based on their suitability and readiness for current and developing computational architectures. Lastly, the greatest opportunities for future research and application are highlighted, including gaps in modeling capabilities, opportunities for near-term application, and key scientific challenges.

Computational science and engineering are shifting toward data-centric, optimization-based, and self-correcting solvers with artificial intelligence. This transition faces challenges such as low accuracy with sparse data, poor scalability, and high computational cost in complex system design. This work introduces Interpolating Neural Network (INN)-a network architecture blending interpolation theory and tensor decomposition. INN significantly reduces computational effort and memory requirements while maintaining high accuracy. Thus, it outperforms traditional partial differential equation (PDE) solvers, machine learning (ML) models, and physics-informed neural networks (PINNs). It also efficiently handles sparse data and enables dynamic updates of nonlinear activation. Demonstrated in metal additive manufacturing, INN rapidly constructs an accurate surrogate model of Laser Powder Bed Fusion (L-PBF) heat transfer simulation. It achieves sub-10-micrometer resolution for a 10 mm path in under 15 minutes on a single GPU, which is 5-8 orders of magnitude faster than competing ML models. This offers a new perspective for addressing challenges in computational science and engineering.

Braking systems are essential to aircraft safety and operational efficiency; however, the variability of carbon brake wear, driven by the intricate interplay of operational and environmental factors, presents challenges for effective maintenance planning. This effort leverages machine learning classifiers to predict wear severity using operational data from an airline’s wide-body fleet equipped with wear pin sensors that measure the percentage of carbon pad remaining on each brake. Aircraft-specific metrics from flight data are augmented with weather and airport parameters from FlightAware® to better capture the operational environment. Through a systematic benchmarking of multiple classifiers, combined with structured hyperparameter tuning and uncertainty quantification, LGBM and Decision Tree models emerge as top performers, achieving predictive accuracies of up to 98.92%. The inclusion of environmental variables substantially improves model performance, with relative humidity and wind direction identified as key predictors. While machine learning has been extensively applied to predictive maintenance contexts, this work advances the field of brake wear prediction by integrating a comprehensive dataset that incorporates operational, environmental, and airport-specific features. In doing so, it addresses a notable gap in the existing literature regarding the impact of contextual variables on brake wear prediction.

Predictive maintenance in commercial aviation demands highly reliable and robust models, particularly for critical components like carbon brakes. This paper addresses two primary concerns in modeling carbon brake wear for distinct aircraft variants: (1) the choice between developing specialized models for individual aircraft types versus a unified, general model, and (2) the potential of transfer learning (TL) to boost model performance across diverse domains (e.g., aircraft types). We evaluate the trade-offs between predictive performance and computational efficiency by comparing specialized models tailored to specific aircraft types with a generalized model designed to predict continuous wear values across multiple aircraft types. Additionally, we explore the efficacy of TL in leveraging existing domain knowledge to enhance predictions in new, related contexts. Our findings demonstrate that a well-tuned generalized model supported by TL offers a viable approach to reducing model complexity and computational demands while maintaining robust and reliable predictive performance. The implications of this research extend beyond aviation, suggesting broader applications in component predictive maintenance where data-driven insights are crucial for operational efficiency and safety.

A data-free, predictive scientific AI model, Tensor-decomposition-based A Priori Surrogate (TAPS), is proposed for tackling ultra large-scale engineering simulations with significant speedup, memory savings, and storage gain. TAPS can effectively obtain surrogate models for high-dimensional parametric problems with equivalent zetta-scale ($10^{21}$) degrees of freedom (DoFs). TAPS achieves this by directly obtaining reduced-order models through solving governing equations with multiple independent variables such as spatial coordinates, parameters, and time. The paper first introduces an AI-enhanced finite element-type interpolation function called convolution hierarchical deep-learning neural network (C-HiDeNN) with tensor decomposition (TD). Subsequently, the generalized space-parameter-time Galerkin weak form and the corresponding matrix form are derived. Through the choice of TAPS hyperparameters, an arbitrary convergence rate can be achieved. To show the capabilities of this framework, TAPS is then used to simulate a large-scale additive manufacturing process as an example and achieves around 1,370x speedup, 14.8x memory savings, and 955x storage gain compared to the finite difference method with $3.46$ billion spatial degrees of freedom (DoFs). As a result, the TAPS framework opens a new avenue for many challenging ultra large-scale engineering problems, such as additive manufacturing and integrated circuit design, among others.

The operational integrity and performance of aircraft braking systems are paramount to commercial aircraft safety and maintenance planning efficiency. Operational integrity refers to the reliability of the braking systems under various conditions, including the ability to withstand high temperatures and resist wear. High operational integrity means fewer unexpected failures, which leads to improved overall aircraft safety and reduced unplanned maintenance. Brake performance includes the effectiveness of braking during landing and taxiing as well as the system's responsiveness to pilot inputs. This study presents a thorough approach to understanding carbon brake pad degradation by benchmarking a suite of unsupervised Machine Learning (ML) clustering algorithms. The objective is to uncover distinct wear patterns and identify salient features differentiating varying degrees of wear. This effort leverages data that includes aircraft-specific parameters (such as aircraft weight), operational conditions (such as flight duration), and environmental factors (such as static air temperature) along with airport characteristics (such as runway length) observed across an airline's fleet of widebody aircraft variants. A Random Forest classifier is implemented to determine the most influential predictors of wear levels, providing a robust feature importance analysis. Methods including Principal Component Analysis (PCA) and an autoencoder are then leveraged to further reduce the dimensionality of the dataset. Various clustering techniques, including K-Means and Agglomerative Clustering, are considered and benchmarked with varying hyperparameter settings. These methods are applied without the knowledge of pre-assigned wear labels, ensuring an unbiased grouping based on intrinsic data characteristics. The performance of these algorithms is then quantitatively assessed using unsupervised evaluation metrics (e.g., Silhouette score) and supervised metrics (e.g., Rand Index) to gain insights from wear labels derived from the available wear pin parameter from the aircraft data. Categorical labels (i.e., High, Medium, or Low wear) are created by interpolating the wear pin signal and categorizing the degradation per flight into quantiles. The top features identified by Random Forest are then analyzed for differences across clusters. This iterative clustering process helps explain the data's intrinsic structure, revealing the foremost features indicative of brake wear. The findings have the potential to contribute to predictive maintenance strategies by enhancing the understanding of how various operating and environmental conditions impact carbon brake pad degradation.

Integrating experiment and simulation provides invaluable insights into the critical parameters that determine the microstructure of alloys produced by additive manufacturing. Here, the grain structure formation due to solidification during single pass laser scans (mimicking bead-on-plate single tracks) on a 316L stainless steel is studied in situ inside a scanning electron microscope that is directly integrated with a continuous-wave laser. The grain size distribution before melting is used as an initial condition in a coupled phase-field/thermal multiphysics modeling framework. The predicted resolidified microstructures are found to agree favorably with those observed experimentally for multiple laser powers and scan velocities, indicating the validity of the overall model. Grain morphology is analyzed quantitatively, and the top surfaces are compared between the experiments and simulations. Analysis of the three-dimensional grain shapes predicted by the simulations shows that the length of the major axis of the resolidified grains is sensitive to laser power and scan speeds, while the length of the minor axis is not. Furthermore, the preferential alignment of the major axes of the grains depends on the melt pool geometry.

A common trend in simulation-driven engineering applications is the ever-increasing size and complexity of the problem, where classical numerical methods typically suffer from significant computational time and huge memory cost. Methods based on artificial intelligence have been extensively investigated to accelerate partial differential equations (PDE) solvers using data-driven surrogates. However, most data-driven surrogates require an extremely large amount of training data. In this paper, we propose the Convolutional Hierarchical Deep Learning Neural Network-Tensor Decomposition (C-HiDeNN-TD) method, which can directly obtain surrogate models by solving large-scale space-time PDE without generating any offline training data. We compare the performance of the proposed method against classical numerical methods for extremely large-scale systems.

Artificial intelligence (AI) has revolutionized software development, shifting from task-specific codes (Software 1.0) to neural network-based approaches (Software 2.0). However, applying this transition in engineering software presents challenges, including low surrogate model accuracy, the curse of dimensionality in inverse design, and rising complexity in physical simulations. We introduce an interpolating neural network (INN), grounded in interpolation theory and tensor decomposition, to realize Engineering Software 2.0 by advancing data training, partial differential equation solving, and parameter calibration. INN offers orders of magnitude fewer trainable/solvable parameters for comparable model accuracy than traditional multi-layer perceptron (MLP) or physics-informed neural networks (PINN). Demonstrated in metal additive manufacturing, INN rapidly constructs an accurate surrogate model of Laser Powder Bed Fusion (L-PBF) heat transfer simulation, achieving sub-10-micrometer resolution for a 10 mm path in under 15 minutes on a single GPU. This makes a transformative step forward across all domains essential to engineering software.

Abstract Challenge 3 of the 2022 NIST additive manufacturing benchmark (AM Bench) experiments asked modelers to submit predictions for solid cooling rate, liquid cooling rate, time above melt, and melt pool geometry for single and multiple track laser powder bed fusion process using moving lasers. An in-house developed A dditive M anufacturing C omputational F luid D ynamics code (AM-CFD) combined with a cylindrical heat source is implemented to accurately predict these experiments. Heuristic heat source calibration is proposed relating volumetric energy density (ψ) based on experiments available in the literature. The parameters of the heat source of the computational model are initially calibrated based on a Higher Order Proper Generalized Decomposition- (HOPGD) based surrogate model. The prediction using the calibrated heat source agrees quantitatively with NIST measurements for different process conditions (laser spot diameter, laser power, and scan speed). A scaling law based on keyhole formation is also utilized in calibrating the parameters of the cylindrical heat source and predicting the challenge experiments. In addition, an improvement on the heat source model is proposed to relate the Volumetric Energy Density (VED σ ) to the melt pool aspect ratio. The model shows further improvement in the prediction of the experimental measurements for the melt pool, including cases at higher VED σ . Overall, it is concluded that the appropriate selection of laser heat source parameterization scheme along with the heat source model is crucial in the accurate prediction of melt pool geometry and thermal measurements while bypassing the expensive computational simulations that consider increased physics equations.

A digital twin (DT) is a virtual representation of physical process, products and/or systems that requires a high-fidelity computational model for continuous update through the integration of sensor data and user input. In the context of laser powder bed fusion (LPBF) additive manufacturing, a digital twin of the manufacturing process can offer predictions for the produced parts, diagnostics for manufacturing defects, as well as control capabilities. This paper introduces a parameterized physics-based digital twin (PPB-DT) for the statistical predictions of LPBF metal additive manufacturing process. We accomplish this by creating a high-fidelity computational model that accurately represents the melt pool phenomena and subsequently calibrating and validating it through controlled experiments. In PPB-DT, a mechanistic reduced-order method-driven stochastic calibration process is introduced, which enables the statistical predictions of the melt pool geometries and the identification of defects such as lack-of-fusion porosity and surface roughness, specifically for diagnostic applications. Leveraging data derived from this physics-based model and experiments, we have trained a machine learning-based digital twin (PPB-ML-DT) model for predicting, monitoring, and controlling melt pool geometries. These proposed digital twin models can be employed for predictions, control, optimization, and quality assurance within the LPBF process, ultimately expediting product development and certification in LPBF-based metal additive manufacturing.

The National Advanced Driving Simulator is a high-fidelity motion-base simulator owned by the National Highway Transportation Safety Administration and managed and operated by the University of Iowa. Its 25-year history has intersected with some of the most significant developments in automotive history, such as advanced driver assistance systems like stability control and collision warning systems, and highly automated vehicles. The simulator is an application of immersive virtual reality that uses multiprojection instead of head-mounted displays. A large-excursion motion system provides realistic acceleration and rotation cues to the driver. Due to its level of immersion and realism, drivers respond to events in the simulator the same way they would in their own vehicle. We document the history and technology behind this national facility.

A key challenge in additive manufacturing (AM) of aluminum-based parts has been the formation of cracks and porosity during the processes. Multicomponent powder beds containing high-melting temperature, highly reactive elements (e.g., Zr and Sc) show promise for improving processibility and reducing crack and pore formation in such alloys. Melting and mixing of these elements in the base alloy during the AM process is yet to be fully understood. This paper describes a multiphysics modeling approach for investigating melt pool dynamics, keyhole and pore formation, and mixing phenomena in multicomponent powder beds. The discrete element method (DEM) is used to generate powder beds with randomly distributed particles of varying sizes. A thermal multi-phase flow model is coupled with a laser welding model in this approach, which includes multiple thermophysical phenomena and laser-material interactions. The multiphysics model was validated using available experimental results in the literature. Through this approach, not only the melt pool dynamics and keyhole morphology, but also the pore formation and mixing evolution during the AM processes, can be quantified for a wide range of process parameters (e.g., laser power and scan speed). To demonstrate the efficacy and application of this method, we thoroughly investigated the additive manufacturing of an Al – Zr powder bed system. The results reveal that the mixing of the alloying element, Zr, is heavily influenced by flow patterns in the melt region and keyhole formation.

Abstract Challenge 3 of the 2022 NIST additive manufacturing benchmark (AM-Bench) experiments asked modelers to submit predictions for solid cooling rate, liquid cooling rate, time above melt, and melt pool geometry for single and multiple track laser powder bed fusion process using moving lasers. An in-house developed Additive Manufacturing Computational Fluid Dynamics code (AM-CFD) combined with a cylindrical heat source was implemented to accurately predict these experiments. Heuristic heat source calibration was proposed relating volumetric energy density (ψ) based on experiments available in the literature. The parameters of the heat source of the computational model were initially calibrated based on a Higher Order Proper Generalized Decomposition- (HOPGD) based surrogate model. The prediction using the calibrated heat source agreed quantitatively with NIST measurements for different process conditions. A scaling law based on keyhole formation was also utilized in calibrating the parameters of the cylindrical heat source and predicting the challenge experiments. In addition, an improvement on the heat source model was proposed to relate the Volumetric Energy Density (VED σ ) to the melt pool aspect ratio. The model showed further improvement in the prediction of the experimental measurements for the melt pool including cases at higher VED σ . Overall, it was concluded that the appropriate selection of parameterization scheme along with the heat source model was crucial in the accurate prediction of melt pool geometry and thermal measurements while bypassing the expensive computational simulations that consider increased physics equations.

Abstract Additive manufacturing (AM) simulations offer an alternative to expensive AM experiments to study the effects of processing conditions on granular microstructures. Existing AM simulations lack support from reliable validation techniques. The stochastic nature and spatial heterogeneity of microstructures make it difficult to validate the simulated microstructures against experimentally obtained images through statistical measures such as average grain size. Another challenge is the lack of reliable and automated methods to calibrate the model parameters, which are unknown and difficult to measure directly from experiments. To overcome these two challenges, we first present a novel metric to quantify the difference between granular microstructures. Then, using this metric in conjunction with Bayesian optimization, we present a framework that can be used to reliably and efficiently calibrate the model parameters. We employ this framework to first calibrate the substrate microstructure simulation and then the laser scan microstructure simulation for Inconel 625. Results show that the framework allows successful calibration of the model parameters in just a small number of simulations.