This study explores the integration of machine learning (ML) with finite element modelling (FEM) to investigate the electrical behaviour of particle-reinforced composites, where a conductive reinforcement phase is dispersed within a non-conductive matrix. Analytical approaches, combined with path-finding algorithms, are employed to determine the effective electrical resistance of the composite. FEM simulations are used to generate datasets representing diverse microstructural configura3ons by systematically varying the particle volume fraction. These datasets form the basis for training ML models, which can then be applied to rapidly predict the effective electrical properties of the composite. This framework enables accelerated material design and discovery.
Ort: C10 | 9.01
Uhrzeit: 14.15 Uhr (und nicht 16.15 Uhr wie gewohnt)