|About this Abstract
||2017 TMS Annual Meeting & Exhibition
||Materials Science for High-Performance Permanent Magnets
||Data-driven Approach for Magnetic Neutron Scattering Data Analysis of Permanent Magnets Using Statistical Learning and Artificial Intelligence
||Kanta Ono, Akinori Asahara, Hidekazu Morita, Chiharu Mitsumata, Masao Yano, Tetsuya Shoji
|On-Site Speaker (Planned)
There has been a growing expectation over recent years for the application of machine learning and artificial intelligence to solve material science issues. Up to now, there is only a few application of artificial intelligence for the analysis of experimental data. Since magnetic small-angle neutron scattering (SANS) is an important technique for characterizing the bulk magnetic microstructure of permanent magnets and provides rich information on the distribution of bulk magnetization on a nanometer length scale, the analysis of the magnetic SANS data is rather complicated. We employed data-driven approach using statistical learning and Artificial Intelligence (Hitachi AI Technology/H) for the analysis of magnetic SANS data and got important aspects of magnetization reversal process of Nd-Fe-B hot-deformed magnets.