Machine learning and predicting the time dependent dynamics of local\n yielding in dry foams

Abstract

The yielding of dry foams is enabled by small elementary yield events on the\nbubble scale, "T1"s. We study the large scale detection of these in an\nexpanding 2D flow geometry using artificial intelligence (AI) and nearest\nneighbour analysis. A good level of accuracy is reached by the AI approach\nusing only a single frame, with the maximum score for vertex centered images\nhighlighting the important role the vertices play in the local yielding of\nfoams. We study the predictability of T1s ahead of time and show that this is\npossible on a timescale related to the waiting time statistics of T1s in local\nneighborhoods. The local T1 event predictability development is asymmetric in\ntime, and measures the variation of the local property to yielding and\nsimilarly the existence of a relaxation timescale post local yielding.\n