Research On Flatness Control System For A Sendzimir Mill

Flatness is an important target for the quality of the strip steel. And the flatness control is the key technology of modern rolling mills with high precision of flatness. Taking the flatness control system for a Sendzimir mill as the research object and combined with the Sendzimir mill cold-rolled stainless steel production line project of a stainless steel plant, a lot of research has been done on the flatness control system for a Sendzimir mill in a stainless steel plant.The structure and characteristics of the Sendzimir mills were analyzed. There were two main kinds of flatness adjustment means in Sendzimir mills. One was the ASU radial adjustment of backup rolls, the other was the axial adjustment of the first intermediate rolls. Researching on the mechanical structure and the control characteristics of the flatness adjustment means provided the basis for making a proper use of the flatness adjustment means and designing of the flatness control system.The static and dynamic models of the Sendzimir mills were builded. Taking into account the strong coupling in the static model, two traditional decoupling methods which are widely used in industry were studied. One was the singular value decomposition decoupling method and the other was the unit matrix decoupling method. And the two decoupling methods were applied in the flatness control system for a Sendzimir mill.The flatness decoupling control method in the flatness control programs of a Sendzimir mill was digested and studied. Taking into account that the ASU radial adjustment of backup rolls is suitable to control low-order flatness defects and the axial adjustment of the first intermediate rolls is suitable to control high-order flatness defects, and also considering the coupling between the two adjustment means, the flatness control system using the unit matrix decoupling method mentioned in chapter4was improved. Theoretical analysis showed that the flatness control system was decoupled into four identical SISO systems. Though MATLAB simulation, it showed that the improved flatness control system effectively reduced the coupling among the multiple loops and improved the quality of the flatness. The method giving here is a valid way in the flatness control system of Sendzimir mills, which gives an idea for other multi-variable coupling systems.Some other parts of the flatness control program were studied, including the signal detection and processing of the flatness, the selection of the flatness control modes, the saturation detection for the actuators and the flatness HMI screen.