1 - 10 of 22 Results
ORNL researchers have developed a deep learning-based approach to rapidly perform high-quality reconstructions from sparse X-ray computed tomography measurements.
Dual-GP addresses limitations in traditional GPBO-driven autonomous experimentation by incorporating an additional surrogate observer and allowing human oversight, this technique improves optimization efficiency via data quality assessment and adaptability to unanticipated exp
We have been working to adapt background oriented schlieren (BOS) imaging to directly visualize building leakage, which is fast and easy.
The lack of real-time insights into how materials evolve during laser powder bed fusion has limited the adoption by inhibiting part qualification. The developed approach provides key data needed to fabricate born qualified parts.
A new nanostructured bainitic steel with accelerated kinetics for bainite formation at 200 C was designed using a coupled CALPHAD, machine learning, and data mining approach.
The invention introduces a novel, customizable method to create, manipulate, and erase polar topological structures in ferroelectric materials using atomic force microscopy.
Scanning transmission electron microscopes are useful for a variety of applications. Atomic defects in materials are critical for areas such as quantum photonics, magnetic storage, and catalysis.
A human-in-the-loop machine learning (hML) technology potentially enhances experimental workflows by integrating human expertise with AI automation.
This work seeks to alter the interface condition through thermal history modification, deposition energy density, and interface surface preparation to prevent interface cracking.
Additive manufacturing (AM) enables the incremental buildup of monolithic components with a variety of materials, and material deposition locations.