Abstract
The dimensional quality of cold rolled sheets is very important for its usage in automotive, packaging and white goods industries. To predict strip shape a partial least square (PLS) regression model has been developed and validated with actual plant data. The model was then used to optimise the process parameters to reduce the magnitude of shape errors. The most important parameters responsible for shape defects have been identified from a large number of input process parameters. With the help of the model it has been found that apart from the tensions and the speeds at different stands, the levelling of the last stand plays an important role in controlling the strip shape. The model can be used online, and is useful for plant operation.
Get full access to this article
View all access options for this article.
