Master Thesis: »Machine Learning (ML)-Based Methods as Surrogate for Finite Element Modelling«

The »« department develops technologies and application-oriented solutions for machining along the entire process chain - from process design and process simulation to real-time data acquisition during production, consulting, and prototype manufacturing. Graph neural networks provide an opportunity to operate on Mesh structured data utilized in Finite Element Method (FEM) simulations and offer time-saving benefits. We are looking for a dedicated and motivated student to assist us in implementing a novel Graph Neural Network based algorithm that can act as surrogate for FEM and accelerate process stability calculation for machining process.