Argonne National Laboratory

Upcoming Events

Data Challenges in Ultrafast X-Ray Imaging: Additive Manufacturing and Beyond

APS Scientific Computation Seminar
Jin Wang (XSD)
May 17, 2018 1:00PM to 2:00PM
Building 401, Room A1100

Abstract: At the Advanced Photon Source (APS), ultrafast and high-speed X-ray imaging and scattering on 100-ps temporal scales has brought a revolutionary tool to scientific and engineering communities interested in studying highly transient phenomena. One applications of thisunique imaging technique is visualizing additive manufacturing (AM) processes.

Additive manufacturing, or 3-D printing, refers to a suite of transformative technologies that build 3-D objects by adding materials layer by layer based on a digital design. Metal AM has found many applications in the biomedical, aerospace, automobile, and defense fields. Despite the many applications, precise control of microstructures and properties of AM products remains challenging owing to the extreme thermal conditions involved in metal AM.

At the APS, we are tackling this problem by using high-speed X-ray imaging for in situ probing of metal AM processes. We demonstrated that important physical processes, including melt pool dynamics, powder spattering ejection, and rapid solidification, can be studied quantitatively with unprecedented spatial and temporal resolution.

On the other hand, the ultrafast and high-speed imaging method has been applied to understand highly transient phenomena in turbulent and cavitating liquid flows. The challenging problem is how to visualize highly transient 3-D morphology from 2-D images in real space and how to combine theory and simulation with experiment data to aid in processing big data sets. With more user groups joining the beamline experiments, we are receiving an ever-increasing demand for effective and efficient computational algorithms for image processing and data analysis.

In this presentation, we will introduce the ultrafast and high-speed experiments at the respective beamlines, explain the challenges in extracting quantitative information from a large volume of X-ray imaging data, and highlight opportunities to use machine learning in understanding dynamic processes.