Publication:
A Hybrid model for hot rolling pass schedule design using reinforcement learning

cris.customurl 15311
dc.contributor.author Kemmerling, Marco
dc.contributor.author Frahm, Nils
dc.contributor.author Idzik, Christian
dc.contributor.author Abdelrazeq, Anas
dc.contributor.author Bailly, David
dc.contributor.author Schmitt, Robert H.
dc.date.issued 2024-03
dc.description.abstract Hot rolling of metals requires the design of pass schedules to accommodate different products and arrive at the desired target properties. Pass schedule design is often still performed manually, although automated approaches are emerging in the literature. One such approach is the use of reinforcement learning, which requires extensive training on simulations. To allow for efficient development of suitable reinforcement learning agents, it is important to reduce the computational cost of each simulation step as much as possible. While some simulation models for hot rolling exist, they have not been developed explicitly for reinforcement learning purposes. Today, there is a lack of fast approximate models to allow for efficient (pre-)training of reinforcement learning agents. To address this gap, we propose a hybrid model consisting of a data-driven model augmented by physical approximations. Our hybrid model produces better predictions than either of its components and is efficient enough to allow for quick training of reinforcement learning agents. We demonstrate that reinforcement learning agents pre-trained on our hybrid model can be transferred to a more accurate simulation and achieve good results after a short additional training phase. The two-stage training approach significantly reduces the overall training while achieving similar agent performances. The reduction in training time enables faster iteration cycles in agent and environment design, thereby creating a foundation for the development of reinforcement learning approaches for a wider range of hot rolling scenarios in the future.
dc.description.version VoR
dc.identifier.doi 10.24405/15311
dc.identifier.uri https://openhsu.ub.hsu-hh.de/handle/10.24405/15311
dc.language.iso en
dc.publisher UB HSU
dc.relation.conference ML4CPS – Machine Learning for Cyber-Physical Systems
dc.relation.orgunit RWTH Aachen
dc.relation.orgunit Institute of Metal Forming, RWTH Aachen
dc.rights.accessRights open access
dc.subject Hot rolling
dc.subject Reinforcement learning
dc.subject Hybrid mode
dc.subject Pass schedule
dc.title A Hybrid model for hot rolling pass schedule design using reinforcement learning
dc.type Conference paper
dcterms.bibliographicCitation.booktitle Machine learning for cyber physical systems
dcterms.bibliographicCitation.originalpublisherplace Hamburg
dcterms.isPartOf https://openhsu.ub.hsu-hh.de/handle/10.24405/16610
dspace.entity.type Publication
hsu.uniBibliography
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
openHSU_15311.pdf
Size:
1.28 MB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
145 B
Format:
Item-specific license agreed upon to submission
Description: