Digitalization for efficient process selection and design of hybrid structures based on experimental and synthetic data

Project runtime: 01.01.2023 – 31.12.2025

Digital manufacturing of components from multiple materials

Hybrid materials, also known as multi-material combinations, are increasingly being used today in mechanical engineering, aerospace and the automotive sector. The material combinations are just as diverse as the joining techniques. In addition to adhesive bonding, mechanical bonding is also widely used. The properties of hybrid materials depend strongly on the joining technique and the process parameters. A systematic survey does not exist at present. Trial-and-error" dominates the development of new hybrid structures. HybridDigital makes a significant contribution to the digitalized and sustainable establishment of hybrid structures for lightweight construction. With the help of a digital twin for hybrids, a resource-optimized and robust development of hybrid structures based on individualized joining is to be possible in the future. The research project focuses on the determination, systematization, structuring, modeling and ultimately formal language description of the process-dependent characteristic values. The data collection is based on the characterization and description of hybrid structures on an experimental and numerical level for a selected multi-material combination (steel-carbon fiber composite, steel-CFRP for short). The necessary knowledge is obtained through systematic specimen fabrication and characterization of the behavior under load, comparatively for two joining techniques (adhesive bonding and bolting), supplemented by acoustic emission analyses. In this way, damage in the material can already be detected at the micro level and related to the failure behavior in the hybrid at the macro level. This allows the properties of the hybrid materials to be linked to the process parameters and the properties of the respective input materials. This should enable the selection of the appropriate joining technique as well as the design of the joint and the automated evaluation of failure and damage patterns. A further added value is provided by the numerical modeling of the scatter of material properties due to material variations and systematic machine errors. As a result, property fluctuations and thus the failure behavior can be better predicted and taken into account accordingly in the component design. Adapted safety factors can reduce the amount of material used and the weight in the component, thus making a direct contribution to the conservation of resources and the environment.

Project presentation unofficial get together (24.04.23): Project presentation HybridDigital

Tasks within the project Location
Cotesa GmbH
Determination of characteristic values and systematic data preparation
Novicos GmbH
Models for numerical analysis
Enari GmbH
Machine learning and connection to the PMD
Garching / Munich
Fraunhofer Institute for Casting, Composite and Processing Technology IGCV and Fraunhofer Institute for Digital Media Technology IDMT
Machine learning and ontologies
Aalen University
Microstructural and mechanical characterization