Digital Twins in deep drawing for virtual tool commissioning and inline parameter optimization

Research output: Contribution to journalResearch articleContributedpeer-review



Fluctuating boundary conditions and the highly nonlinear process behavior of deep drawing operations make experience-based selection of suitable control parameters difficult. Nowadays, commissioning deep drawing tools, i.e., die spotting and identification of suitable control parameters for defect-free parts, is conducted on real world try-out presses. Transferring the tools to production machines entails adaption of these initial parameters. Once production is ramped up, any changes to material properties, lubrication and press behavior require continuous manual machine parameter tuning. Virtual tool commissioning and utilizing these simulation models in the production phase to adapt control parameters would reduce time and cost over the entire life cycle of the machine and the tool sets. The virtual representative, which provides services for a plant over several life cycle phases, is also referred to as Digital Twin. In this paper, the authors present a Digital Twin concept for deep drawing presses to predict the state of the system and optimize the control parameters during the production. The integration of all involved subsystems into one system simulation and its efficient calculation is the biggest challenge. The authors combine a virtual commissioning simulation tool with a finite element model to implement all relevant properties of the deep drawing press and the interaction of its subsystems. It is shown, how the idea of a system simulation makes predictions of system parameters specific to a production situation possible, and therefore, can help to select suitable control parameters that leads to a reduction of the error rate in deep drawing.


Original languageEnglish
Article number012072
Number of pages9
JournalIOP Conference Series: Materials Science and Engineering
Issue number1
Publication statusPublished - 1 May 2022

External IDs

Mendeley c5cfa6d0-afb6-333b-b4ef-7ebe657ad3d4
unpaywall 10.1088/1757-899x/1238/1/012072
WOS 000894042400072


Library keywords