Engineering Papers⌕ Search

Engineering topics

Peles, Amra

Publications and source records attributed to Peles, Amra.

Deep-learning based artificial intelligence tool for melt pools and defect segmentation

Accelerating fabrication of additively manufactured components with precise microstructures is important for quality and qualification of built parts, as well as for a fundamental understanding of process improvement. Accomplishing this requires fast and robust characterization of melt pool geometries and structural defects in images. This paper proposes a pragmatic approach based on implementation of deep learning models and self-consistent workflow that enable systematic segmentation of defects and melt pools in optical images. Deep learning is based on an image-to-image translation–conditional generative adversarial neural network architecture. An artificial intelligence (AI) tool based on this deep learning model enables fast and incrementally more accurate predictions of the prevalent geometric features, including melt pool boundaries and printing-induced structural defects. We present statistical analysis of geometric features that is enabled by the AI tool, showing strong spatial correlation of defects and the melt pool boundaries. The correlations of widths and heights of melt pools with dataset processing parameters show the highest sensitivity to thermal influences resulting from laser passes in adjacent and subsequent layer passes. The presented models and tools are demonstrated on the aluminum alloy and datasets produced with different sets of processing parameters. However, they have universal quality and could easily be adapted to different material compositions. The method can be easily generalized to microstructural characterizations other than optical microscopy.

additive manufacturing↗

A Co-Registered In-Situ and Ex-Situ Dataset from a Laser Powder Bed Fusion Additive Manufacturing Process (Peregrine v2023-10)

This release contains a co-registered in-situ and ex-situ Peregrine dataset from a single Concept Laser M2 Laser Powder Bed Fusion (L-PBF) stainless steel 316L build. These data were collected at the Manufacturing Demonstration Facility (MDF) located at Oak Ridge National Laboratory (ORNL). The dataset includes layer-wise visible-light in-situ imaging data, the laser scan paths and parameters, in-situ temporal sensor data, X-Ray Computed Tomography (X-CT) scans, pycnometry and tensile test results, etched micrographs from selected locations, and the target part geometries. Additionally, anomaly detections produced by a modified Dynamic Segmentation Convolutional Neural Network (DSCNN) are provided.

36 MATERIALS SCIENCE↗

Report Outlining Computed Tomography Strategy and Microscopy Approach to Qualifying AM 316 Materials

This report is part of work package CR-22OR0406012, Automated, High-Throughput Materials Characterization Techniques , under the Advanced Materials and Manufacturing Technologies program (AMMT). The project’s primary objective is to leverage our AI-based rapid and high-throughput automated characterization framework to qualify additively manufactured 316 materials comprehensively, focusing on optimizing the additive manufacturing process and evaluating the performance of 3D-printed stainless steel components. This report outlines our strategy for leveraging the automated characterization process for qualifying 316H materials.

36 MATERIALS SCIENCE↗

Sparse Data Machine Learning Integration with Theory, Experiment and Uncertainty Quantification: Process-Structure-Property-Performance of Friction Deformation Processing

Computer vision and deep learning tools that advance the ability to establish processing-structure-property-performance (PSPP) relations are presented. The Bayesian binning method for image segmentation enables quantitative analysis of microstructural features in an automated way, while the analysis of shapes and relative orientation of these features reveals local deformation maps indicative of both, material flow and residual stresses due to materials processing. The deep learning method leads to the previous knowledge agnostic mapping of empirically observed microstructural zones in friction stir welding (FSW) process and synthetic microstructure generation capability that is statistically equivalent to experimentally collected data.

97 MATHEMATICS AND COMPUTING↗

"Sintering" Models and In-Situ Experiments: Data Assimilation for Microstructure Prediction in SLS Additive Manufacturing of Nylon Components

Selective laser sintering methods are workhorses for additively manufacturing polymer-based components. The ease of rapid prototyping also means it is easy to produce illicit components. It is necessary to have a data-calibrated in-situ physical model of the build process in order to predict expected and defective microstructure characteristics that inform component provenance. Toward this end, sintering models are calibrated and characteristics such as component defects are explored. This is accomplished by assimilating multiple data streams, imaging analysis, and computational model predictions in an adaptive Bayesian parameter estimation algorithm. From these data sources, along with a phase-field model, bulk porosity distributions are inferred. Model parameters are constrained to physically-relevant search directions by sensitivity analysis, and then matched to predictions using adaptive sampling. Using this feedback loop, data-constrained estimates of sintering model parameters along with uncertainty bounds are obtained.

3D printing, Additive Manufacturing, polymer, sint↗

GGA+U method from first principles: application to reduction–oxidation properties in ceria-based oxides

We implement DFT calculations by a Hubbard-like correction for localized strongly correlated electrons, coupled with a generalized gradient approximation to the exchange-correlation functional to elucidate the role of the ceria based catalytically active supports for the chemical reactions involving reduction–oxidation processes. These catalytic processes are relevant for many industrial applications, such as catalytic converters in automotive applications, solid oxide fuel cells, and hydrogen production from biomass. The Hubbard-like correction U is computed from first principles as physical property of the system. We find that the high performance of ceria-based oxides as an active support for noble metals in catalysis relies on an efficient supply of lattice oxygen at reaction sites governed by oxygen vacancy formation.

Materials Science↗

Electrocatalysis on Platinum Nanoparticles: Particle Size Effect on Oxygen Reduction Reaction Activity

We determined the size-dependent specific and mass activities of the oxygen reduction in HClO4 solutions on the Pt particles in the range of 1–5 nm. The maximal mass activity at 2.2 nm is well explained based on density functional theory calculations performed on fully relaxed nanoparticles. The presence of the edge sites is the main reason for the low specific activity in nanoparticles due to very strong oxygen binding energies at these sites. Our results clearly demonstrate that the catalytic activity highly depends on the shape and size of the nanoparticles.

Chemistry↗

Enhanced Oxygen Reduction Activity of Platinum Monolayer on Gold Nanoparticles

The increase in oxygen binding energy was previously proposed to account for the lower oxygen reduction activity of a Pt monolayer supported on Au(111) single crystal than that on Pd(111) and pure Pt(111) surfaces. This single-crystal based understanding, however, cannot explain the new finding of a 1.6-fold increase of oxygen reduction activity on Pt monolayer-modified 3-nm Au nanoparticles (Pt/Au/C) in comparison with that on Pt/Pd/C with a similar particle size. The Pt/Au/C catalyst also has an activity higher than that of a state-of-the-art 2.8-nm Pt/C catalyst. Our new experimental results and density functional theory calculations demonstrate that a significant compressive strain in the surface of the core nanoparticles plays a role in the observed activity enhancement.

Chemistry↗

Pt Monolayer on Porous Pd-Cu Alloys as Oxygen Reduction Electrocatalysts †

We demonstrate the synthesis of a core-shell catalyst consisting of a Pt monolayer as the shell and porous/hollow Pd-Cu alloy nanoparticles as the core. The porous/hollow Pd-Cu nanoparticles were fabricated by selectively dissolving a less noble metal, Cu, using an electrochemical dealloying process. The Pt mass activity for the oxygen reduction reaction of a Pt monolayer deposited on such a porous core is 3.5 times higher than that of a Pt monolayer deposited on bulk Pd nanoparticles and 14 times higher than that of state-of-the-art Pt/C electrocatalysts.

Chemistry↗