Roll diameter surface deviations can generate significant strip flatness defects in the cold rolling of thin gauge metal sheet. The resolution of roll profile deviations can vary between 0.001 and 0.01 mm. In such a high-fidelity contact problem, lack of availability of the measured data, and high computational cost in using the data in simulations, make it very difficult to predict the effects of such deviations. Nonetheless, an understanding of how high-fidelity roll profile deviations can lead to rolled sheet flatness defects is very important since such defects may be unrecoverable. This paper evaluates the ability of a simplified mixed finite element roll-stack model to predict contact behavior effects on the rolled sheet created by high-fidelity roll profile deviations. The simplified finite element method combines 3D Timoshenko beam elements with Winkler elastic foundations. The beam elements are used to capture bending and shear deformation, while the Winkler foundations are tested for their ability to accurately and efficiently capture the high-fidelity flattening-type contact behavior. Results for a simple roll and plate contact case study indicate favorable comparison with the results obtained from a large-scale commercial finite element simulation, and yet the methods requires a small fraction of the associated computing time and memory. The work also offers significant insights into the sheet flatness defects that can arise in cold rolling because of low magnitude but high fidelity roll diameter machining errors.
High-Fidelity Roll Profile Contact Modeling by Simplified Mixed Finite Element Method
- Views Icon Views
- Share Icon Share
- Search Site
Zhang, F, Malik, AS, & Yu, H. "High-Fidelity Roll Profile Contact Modeling by Simplified Mixed Finite Element Method." Proceedings of the ASME 2018 13th International Manufacturing Science and Engineering Conference. Volume 4: Processes. College Station, Texas, USA. June 18–22, 2018. V004T03A034. ASME. https://doi.org/10.1115/MSEC2018-6509
Download citation file: