.
Performance optimization in the large format metal additive manufacturing (MAM) of ultrahigh strength steels (UHSS) is required to take advantage of increasingly complex mechanical component designs. Previous process-microstructure-property (PMP) relationship development led to crack-free single pass wall builds. However, hot cracking was found in solute-rich, interdendritic regions of the fusion zone and weld metal heat-affected zone (HAZ) in thicker section multipass builds. Continuing process optimization required identification of the cracking mechanisms. This work analyzed the cracking mechanisms through metallurgical characterization and computational thermodynamics calculations. Solidification and potential weld metal liquation cracking were related to the non-equilibrium solidification conditions, causing non-equilibrium partitioning of solute elements to solidification grain boundary (SGB) and solidification subgrain boundary (SSGB) regions during DED-Arc fabrication. Solidification cracking was caused by non-equilibrium partitioning of carbon, changing the solidification mode from ferritic to austenitic and expanding the solidification temperature range (STR). Expansion of the STR was compounded by partitioning of impurities, phosphorus and sulfur, during austenitic solidification, forming persistent, low solidification temperature liquid films. Tensile stresses induced by solidification shrinkage and thermal contraction during DED-Arc fabrication restrained the solidifying material, resulting in cracks at SGB and SSGB regions. Weld metal liquation cracking conditions were created by reheating of low melting temperature, solute-rich constituents at SGBs and SSGBs within HAZs of subsequently deposited DED-Arc beads. Formation of a mixed martensite and austenite microstructure was related to solute partitioning in the non-equilibrium solidification conditions of DED-Arc. Reheating during fabrication and the final stress relief heat treatment affected the solid-state transformations.
Modern engineering and scientific workflows often require simultaneous predictions across related tasks and fidelity levels, where high-fidelity data is scarce and expensive, while low-fidelity data is more abundant. This paper introduces an Multi-Task Gaussian Processes (MTGP) framework tailored for engineering systems characterized by multi-source, multi-fidelity data, addressing challenges of data sparsity and varying task correlations. The proposed framework leverages inter-task relationships across outputs and fidelity levels to improve predictive performance and reduce computational costs. The framework is validated across three representative scenarios: Forrester function benchmark, 3D ellipsoidal void modeling, and friction-stir welding. By quantifying and leveraging inter-task relationships, the proposed MTGP framework offers a robust and scalable solution for predictive modeling in domains with significant computational and experimental costs, supporting informed decision-making and efficient resource utilization.
Consumer, industrial, and government products contain many different materials to decrease weight, increase efficacy, and present new functionalities. These include health monitoring devices, golf clubs, and other sporting equipment for consumer goods, eVTOLs, drones, and aircraft for aerospace, and electric, hybrid, or gas vehicles for automotive. Dissimilar material joints, or mixed-material composites, can be difficult to separate at the end of the products’ lives. A profound understanding of the molecular structure at interfaces is required for joints to ensure the product will last its intended life and be detachable. During this presentation lightweight, strong, and reliable joint designs will be described for both joining and end-of-life separation. One example is polymer-to-metal direct joining where heat is used for the joint assembly and disassembly so that the polymer and metal can easily be sorted into different recovered streams for reuse. Other topics during the presentation will include material selection and joint design, which are equally as important for a successful product with materials that are poised for recovery.
Constitutive modeling of the relationship between process-imposed material states and fundamental material properties is critical to control of material microstructure in manufacturing processes. The limited accuracy resulting from the typical reliance on fallible human expertise and intuition for postulation and revision of the models functional form results in incremental and time consuming model discovery. Conventional Machine Learning (ML) incurs significant cost and time of data generation. Model discovery using Large Language Models (LLMs) suffers from the above issues and/or ignores the inviolability of fundamental thermodynamics laws. This work creates a novel GPT-Micro paradigm for autonomous, data sparse, and thermodynamics-compliant discovery of de-novo constitutive models. This framework seamlessly integrates semantic knowledge extraction from literature, enforcement of thermodynamics-based conservation laws, and sparse datasets, with LLM-driven generation and refinement of model hypotheses. Validation is performed for a long-intractable constitutive modeling problem in a printed electronics process testbed. This reveals significant and simultaneous advantages over the state-of-the-art including: (a) More than 70 percent reduction in data burden relative to ML-based modeling without loss in accuracy; (b) 400X reduction in discovery time after data generation, from months to hours, relative to human-driven modeling; (c) Discovery of models with novel functional forms without subjective human choice of a starting hypothesis; (d) Enhanced physics-rooted trustworthiness, human interpretability, and mechanistic insight via synthesis of compact, conservation-compliant, and physically complete analytical models. The potential of GPT-Micro to realize rapid, low-cost, physically trustworthy, and interpretable microstructure modeling across the manufacturing landscape is discussed.
Additive manufacturing and welding processes are highly sensitive to heat dissipation, where improper thermal management leads to residual stresses, distortions, and cracking. Existing heat transfer models, such as Rosenthal's solutions, fail to handle finite 3D geometries, cooling effects, or transient behavior, limiting their accuracy. We overcome these limitations by developing an analytical framework that incorporates cooling boundary conditions mimicking Newton's Law of Cooling. Using two different and proven-equivalent approaches, Laplace transform and Fourier series, we derive closed-form solutions for transient and steady-state temperature profiles under various heat sources, including Gaussian, ellipsoidal, double-ellipsoidal, and time-dependent on/off switch sources. We compare our analytical solutions to numerical implementations, demonstrating strong agreement while providing deeper physical insight. This approach significantly reduces computational cost and experimental requirements, making it a scalable tool for optimizing thermal predictions and mitigating residual stresses in metal-based manufacturing. Additionally, our framework enables the generation of synthetic datasets for machine learning models to predict heat distribution efficiently.
