ATAP researchers discuss tests of a technique that monitors an entire magnet for safe temperature conditions in order to avoid quenches before they occur. AI/ML is emerging as an important tool in analyzing diagnostic data in search of quench precursors. It is expected to ultimately serve in realtime operational decision making.

New diagnostics and instrumentation are being developed to glean deeper insight into the performance limitations of high field field superconducting magnets and to support their protection in the case of local loss of superconductivity.

The data flow from these diagnostics is significant, and analyzing the data to extract critical physics is a major challenge. AI/ML is well suited to help in this endeavor, as it can identify hidden correlations, help identify weak but critical signals, and ultimately support real-time decision-making processes to protect magnets.

Characterization of commercial conductors are important parts of our work in the Superconducting Magnet Program. AI/ML techniques are proving tremendously valuable in processing microscopy data.

At the superconductor level, scientists from universities, national laboratories, and industry have long utilized advanced microscopy techniques such as scanning and transmission electron microscopes (SEM/TEM) to characterize conductors. AI/ML image analysis techniques are proving to be extremely valuable in automating the characterization and in identifying critical features through statistical correlations contained in the large data sets of microscopy images of conductors, both currently and moving forward.