| About this Abstract |
| Meeting |
TMS Specialty Congress 2025
|
| Symposium
|
8th World Congress on Integrated Computational Materials Engineering (ICME 2025)
|
| Presentation Title |
Forecasting of Roll Loads During Hot Rolling Using an Artificial Neural Network Model in a Hot-Strip Mill |
| Author(s) |
Abhishek Kumar Thakur, Sanyam Nitin Totade, Appa Rao Chintha |
| On-Site Speaker (Planned) |
Abhishek Kumar Thakur |
| Abstract Scope |
Hot rolling is an important metallurgical process in which a hot steel slab is subjected to thickness reduction by passing through a series of rolls in a hot-strip mill. During hot rolling, the rolls at each stand experiences a tremendous amount of force which depends on the hot metal chemistry and process parameters. The forecast of these roll loads can provide a guide and help optimizing the overall hot rolling process. In this context, we developed an artificial neural network model which can predict the rolling loads for a given set of steel chemistry and process parameters. The network is trained using industrial data with compositions and process parameters used as input and roll loads predicted as output. During testing, on an average, the trained neural network model can predict 74.24% and 95.06% of rolling loads within ±5% and ±10% of the corresponding actual roll loads respectively. |
| Proceedings Inclusion? |
Undecided |