Project Details
Projekt
T2: Physics-based process evaluation and decision support for structural process improvement
Applicants
Professor Dr.-Ing. Jürgen Fleischer; Professor Dr.-Ing. Frank Henning; Professorin Dr.-Ing. Luise Kärger
Subject Area
Lightweight Construction, Textile Technology
Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing
Plastics Engineering
Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing
Plastics Engineering
Term
since 2022
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 459291153
Project T2 focuses on a virtual counterpart of the physical process construction kit, in order to further increase the efficiency of process maturation during production ramp-up. Physics-based virtual process evaluation helps to escape the local optima of the physical process layout and to support structural process improvements, via guided exploration of the function-oriented design space. If the improvements based on data-driven model learning and process design are exhausted, the process may still be immature. In this case, physics-based process simulation models can be used – in conjunction with the physical data stored thus far in the knowledge base – to serve as an additional source of (virtual) process data and to provide an understandable and insightful decision support for process experts, in order to allow them to modify the process structure in a purposive way. Thus, physics-based process simulations will be utilised to provide virtual process data that are not accessible from physical sensors, and to explore virtual process configurations beyond the already implemented real process instance. Based on thermomechanical material modelling, high-fidelity process simulation models for stamp forming, demoulding and rework (via SPIF - single-point incremental forming) will be utilized, further developed, and connected to a modular virtual process chain (CAE chain). In addition to high-fidelity, also low-fidelity models will be considered to increase computational efficiency. Approaches for model validation and data assimilation are applied to detect insufficient modelling and parametrisation, and to assess the overall prediction capability of the CAE chain. Finally, structural deficits and suggestions for structural improvements of the process will be deduced, which can then be transferred, specified and implemented in T1 in the physical process construction kit.
DFG Programme
Research Units
