Session Sheet - 2026 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2026): Applications: Residual Stress

2026 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2026): Applications: Residual Stress
Program Organizers: David Leigh, University of Texas at Austin

Tuesday 8:00 AM
August 4, 2026
Room: Brazos
Location: AT&T Center

8:00 AM
In-Situ Characterization of Residual Stresses in Powder-Based Directed Energy Deposition: Andrea Camacho-Betancourt¹; Tor Slowe¹; Iris Rivero¹; ¹University of Florida
    This research investigates the feasibility of the ψ² X-ray diffraction (XRD) method for in-situ, non-destructive residual stress analysis during powder-based directed energy deposition (DED) of 316 stainless steel powder onto a matching substrate. Where conventional techniques, like the contour method, require mechanical post-processing, in-situ XRD enables real-time tracking of lattice strain. Using lateral tilts, the ψ² method obtains measurements without reorienting the sample immediately following deposition. Single-layer depositions were examined to isolate the effects of localized thermal cycles during deposition and subsequently compared to measurements post-cooling after residual stresses profiles stabilized. Residual stresses measurements using the ψ² XRD method enables tracking of the lattice strain evolution, showing thermal cycling and high tensile residual stress in the melt pool region. Ultimately, real-time tracking can provide predictive insight into final the state of residual stresses on the DED fabricated components enabling process parameter optimization for remanufacturing applications.

8:20 AM
Build-Plate-Free Wire-Arc Direct Energy Deposition: Neutron Diffraction and Computational Study: Aslan Nasirov¹; Wen Dong¹; Christopher Fancher¹; Bhagya Prabhune¹; Jerry Anzalone²; Russel Stein²; Srdjan Simunovic¹; Andrzej Nycz¹; Yousub Lee¹; ¹Oak Ridge National Laboratory; ²Granularity, LLC
    Wire-arc Direct Energy Deposition (Wire-arc DED) enables efficient fabrication of large metallic components because of its high deposition rate and high material efficiency but relies on a build plate to constrain deformation during printing. Build-plate-free (“freeform”) approach provides greater geometric flexibility but introduces additional challenges in predicting and controlling residual stresses. This study proposes a simulation framework for predicting residual strain in freeform wire-arc DED and validates the predictions using neutron diffraction measurements. Three wall parts were investigated: (i) an as-deposited wall using conventional 3-axis deposition, (ii) a wall after build-plate removal, and (iii) a 5-axis freeform wall. Residual strains were measured at ORNL’s neutron facility, High Flux Isotope Reactor (HFIR). Sequentially coupled thermo-mechanical finite element simulations, modeling the deposition process, captured the measured residual strain trends for all three walls and demonstrated the capability of the modeling approach to predict residual strain evolution in freeform Wire-arc DED.

8:40 AM
Three Dimensional Semi-Analytical Thermo-Elasto-Plastic Solution for Additive Manufacturing: Tao Liu¹; Ming Leu²; Edward Kinzel³; Chinedum Okwudire¹; ¹University of Michigan; ²Missouri University of Science and Technology; ³University of Notre Dame
    In additive manufacturing (AM), stress drives distortion, warping, delamination, and cracking, making fast and accurate stress prediction essential for process planning, real-time control, digital twins, and data-driven workflows. However, spot-wise thermo-elasto-plastic FEM is often too slow for design iteration or real-time use, while existing semi-analytical approaches are mostly limited to 2D. This paper presents a full 3D semi-analytical thermo-elasto-plastic model for AM. A spot-wise analytical thermal solution is used as the load, and displacements and stresses are computed through Green’s function representations in a half-space. Plastic flow is integrated using an implicit return-mapping algorithm. A surface-consistent singular integral regularization ensures stable and accurate stress evaluation near the traction-free top surface. The model is validated against FEM, showing close agreement in temperature, displacement, and stress fields. Runtime comparisons show up to 1,670× speedup, enabling scalable data generation, scan optimization, and adaptive AM control.

9:00 AM
Towards SmartScan 3.0: A Scan Sequence Optimization Approach for Distortion Reduction in LPBF using Reduced-Order Thermo-Elasto-Plastic Models: Tao Liu¹; Chinedum Okwudire¹; ¹University of Michigan
    In Laser Powder Bed Fusion (LPBF), scan sequence strongly influences residual-stress-driven distortion, yet it is often chosen heuristically because thermo-elasto-plastic models are too costly for large-design-space search. We present a first step towards SmartScan 3.0, an approach that uses a reduced-order thermo-elasto-plastic model (ROM) for scan-sequence optimization. The ROM couples a finite-difference thermal solver with a history-dependent plastic-strain update to predict distortion. ROM predictions are validated against a commercial finite element solver. A search-based optimization over a jump parameter identifies a preliminary SmartScan 3.0 sequence that significantly reduces distortion in laser-marking experiments vis-a-vis SmartScan 1.0 (thermal-model) and 2.0 (thermo-elastic-model) approaches.

9:20 AM
A Transformer-Based Inherent Strain Surrogate Model for Mechanical Field Prediction in Laser Powder Bed Fusion: Shane Garner¹; Praveen Vulimiri¹; Albert To¹; ¹University of Pittsburgh
    Laser powder bed fusion (LPBF) is a widely used additive manufacturing process for producing complex metal components. However, localized thermal gradients during fabrication generate residual stresses that often lead to distortion, cracking, and build failure. While inherent strain simulations can estimate these effects, they are computationally expensive when applied to large-scale or complex design spaces. In this work, a data-driven surrogate model based on a sparse 3D transformer architecture is developed to provide a rapid approximation of such simulations. The model learns the mapping between voxelized geometries and the resulting mechanical fields, including stress, strain, and displacement. By leveraging a local attention mechanism over structured voxel neighborhoods, the model captures both short-range load transfer and long-range mechanical interactions driven by geometry and boundary conditions. The proposed framework significantly reduces computational cost compared to finite element simulations, providing a scalable tool for process-aware design and evaluation of complex LPBF geometries.

9:40 AM Break