Research Seminar on AI: Potentials and Challenges of Data-based Approaches in and around Model-predictive Quality Control
- +49 241 80 28488
- E-Mail schreiben
Dienstag, 30.11.2021, 16.00 Uhr
Model-predictive control is a class of advanced control methods, which enables the operation of production systems at their technological limits, while respecting existing constraints. Thus, model-predictive quality control can be utilized to maximize the production volume without neglecting its implications on the reproducible quality of the product. However, the suggested approach requires adequate knowledge about the behavior of the controlled system and its constraints, which is often aggravated by unknown dependencies and the complexity of rigorous physical models. Data-based approaches paired with shared databases bear the potential to support the model-predictive quality control at discovering and considering unknown behavior. Furthermore, data-based methods can be employed to help configuring the underlying control systems and thus to promote further automation of the production processes. The discussion will be constructed as an excerpt from conceivable data-based methods, regarding their potentials and challenges in and around a model-predictive quality control.
Muzaffer Ay is a research associate with Prof. Dr.-Ing. Abel at the Institute of Automatic Control (IRT) since 2018. He received his B.Sc. degree in Mechanical Engineering and M.Sc. degree in Automation Engineering, both from RWTH Aachen University. His research focuses on applying data-based approaches around the model-predictive quality control of production processes within an “Internet of Production” (IOP), a cluster of excellence at RWTH Aachen.