Fig. 1: A modified 3D Monte Carlo Potts simulation of microstructural coarsening over a period of 100 × 106 simulation steps (100M MCS).
From: Learning to predict rare events: the case of abnormal grain growth
Cross-sections of the simulation are shown every 10M MCS (Monte Carlo steps). The highlighted (red) grain becomes abnormally large (i.e., having a grain volume that is greater than 10 times the initial average grain volume) just after 67M MCS. Our methods predicted that this grain would become abnormal using only the data from 11M to 15M MCS (bracket at left), which is long before abnormality occurs.
