Beschleunigtes Materialdesign durch künstliche Intelligenz im Forschungsdatenmanagement

Abstract

Using the example of polyurethane foam structures, this work develops a modular, FAIR workflow for data-driven materials development. Through AI-based segmentation, generative 3D models, and simulations, microstructural properties are automatically analysed and mechanical characteristics are reliably predicted. The generic workflows and transparent data management accelerate the development process and can be flexibly applied to other materials.