Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
DY: Fachverband Dynamik und Statistische Physik
DY 33: Poster: Nonlinear Dynamics, Pattern Formation and Networks
DY 33.17: Poster
Mittwoch, 20. März 2024, 15:00–18:00, Poster C
Experimental study of stress in force chains in granular matter — •Lukas Reiter1, Amelie Mayländer1, Raphael Blumenfeld3, Clara Wanjura2, and Othmar Marti1 — 1Institute of Experimental Physics, Ulm University, D-89069 Ulm — 2Max Planck Institute for the Science of Light, Staudststr. 7, D-91058 Erlangen — 3Gonville & Caius College, University of Cambridge, Trinity St., Cambridge CB2 1TA, UK
The properties of dense granular media are largely determined by the contact forces between particles. Experimentally, these forces become visible as interference patterns in photo-elastic particles, but, so far, their quantitative analysis from experimental data has been challenging. Using a dark field polariscope, we explore the stress dynamics of a sheared, two-dimensional granular system of photo-elastic discs forming a self-organizing many-particle contact network and observe the formation of force chains. We use a convolutional neural network (CNN) approach based on [1] to analyse the interference fringes arising in the photo-elastic particles due to strain. We train and compare different pre-trained state-of-the-art CNN models on synthetically generated 2D images of particles. The CNNs provide quantitative information on the number of forces, their magnitudes and angles at which the forces are applied.
[1] R. Sergazinov, M. Kramár. Mach. Learn.: Sci. Technol. 2 045030 (2021).
Keywords: granular matter; convolutional neural network; polariscope; force chains; fringes