The team published their findings in the Proceedings of the National Academy of Sciences in an article titled “ Harnessing interpretable and unsupervised machine learning to address big data from modern X-ray diffraction.”
Atomic heart music upgrade#
Along with the upgrade will come a significant increase in data collected at the APS, and machine learning techniques will be essential to analyzing that data in a timely manner. The APS is undergoing a massive upgrade that will increase the brightness of its X-ray beams by up to 500 times. “And our method is applicable to many big data problems in not only superconductors, but also batteries, solar cells, and any temperature-sensitive device.” “Because of machine learning, we are able to see materials’ behavior not visible by conventional XRD,” Osborn said. Those include not only distortions in the orderly arrangement of atoms in the material, but also fluctuations that occur when such changes happen. More important for this study, other unusual features emerge at higher temperatures related to changes in the material structure.īy applying XTEC, the team extracted an unprecedented amount of information about changes in atomic structure at different temperatures. At this ultralow temperature, these materials switch to a superconducting state, offering no resistance to electrical current. The colors tell whether each cluster represents the equivalent of a road, forest or lake in an aerial map.”Īs a test case, XTEC analyzed data from beamline 6-ID-D at the APS, taken from two crystalline materials that are superconducting at temperatures close to absolute zero. “Then, cluster two would be blue, and associated with another property with a different temperature dependence, and so on. “For example, XTEC might assign red to data cluster one, which is associated with a certain property that changes with temperature in a particular way,” Osborn said. These patterns are then represented by color coding. Instead, it learns by finding patterns and clusters in large data sets without such training. This machine learning does not depend on initial training and learning with data already well studied.
+-+Back.jpg "atomic heart music")
XTEC draws on the power of unsupervised machine learning, using methods developed for this project at Cornell University. A typical large data set would be 10,000 gigabytes, equivalent to roughly 3 million songs of streaming music.
It accelerates materials discoveries through rapid clustering and color coding of large X-ray data sets to reveal previously hidden structural changes that occur as temperature increases or decreases. The team calls their new method X-ray Temperature Clustering, or XTEC for short. Sorely lacking, however, are analysis methods that can cope with these immense data sets. In recent decades, the amount of data being produced in XRD experiments has increased dramatically at large facilities such as the Advanced Photon Source ( APS), a DOE Office of Science user facility at Argonne. It has provided key information on the 3D atomic structure of innumerable technologically important materials. “Because of machine learning, we are able to see materials behavior not visible by conventional XRD.” - Raymond Osborn, senior physicistįor over a century, X-ray diffraction, or XRD, has been one of the most fruitful of all scientific methods for analyzing materials. “What might have taken us months in the past, now takes about a quarter hour, with much more fine grained results.” “Our method uses machine learning to rapidly analyze immense amounts of data from X-ray diffraction,” said Raymond Osborn, senior physicist in Argonne’s Materials Science division. It should greatly accelerate future research on structural changes on the atomic scale induced by varying temperature. This new tool uses computational data sorting to find clusters related to physical properties, such as an atomic distortion in a crystal structure. Department of Energy’s ( DOE) Argonne National Laboratory has devised a method for creating color coded graphs of large volumes of data from X-ray analysis. Working with several universities, the U.S. Through color, we can tell at a glance where there is a road, forest, desert, city, river or lake. Color coding makes aerial maps much more easily understood.
0 kommentar(er)
