| About this Abstract |
| Meeting |
13th International Conference on the Technology of Plasticity (ICTP 2020)
|
| Symposium
|
13th International Conference on the Technology of Plasticity (ICTP 2020)
|
| Presentation Title |
Novel Concepts to Optimize Pass Schedules for Rolling in the Context of Ongoing Digitalization |
| Author(s) |
Christian Idzik, Alexander Maximilian Krämer, Johannes Lohmar, Gerhard Hirt |
| On-Site Speaker (Planned) |
Christian Idzik |
| Abstract Scope |
Rolling is a well-established forming process employed in many industrial sectors. Although highly optimized, process disturbances can still lead to undesired properties. This paper demonstrates advances of robust process design based on machine learning, simulated and measured data to minimize scrap and guarantee product quality. Integrating an established physical strengthening model into a rolling model allows tracking the microstructure properties throughout the process, which enables predicting the final properties, e.g. yield strength. A trained machine learning algorithm automatically designs robust pass schedules by implying stricter tolerances with respect to final properties. This technique can prospectively be extended to account for significant disturbances by adapting the process to reach the desired properties and reduce scrap. Finally, concepts of considering the whole process chain when optimizing or adapting each process are presented. In conclusion, the results show the potential of increased digitalization for further quality improvements even in a highly optimized environment. |
| Proceedings Inclusion? |
Definite: At-meeting proceedings |