Fig. 2 Schematic of a virtual rolling mill

MCS is the core of process control system including all logic and continuos process control functions. Mainly composed of two Modicon PLCs, it is connected to the real mill or virtual mill

simulator only through a different interface. MCS main functions therefore would not distinguish between real mill and the VMS.

HMI is an interface between human operator and the process. The operator monitors and controls the process through various functional screens provided by HMI, e.g. manually correcting the stand speeds by observing each motor current.

VMS is to simulate the complex dynamics of a real mill consisting of 5 roughing rolling stands, 3 intermediate rolling stands and 8 finishing rolling stands. Although highly ordered, nonlinear, coupled and imprecise process models are not suitable for control system design, they are very useful in process simulation and numerical analysis.

The information exchange in the virtual rolling mill is completed through a MB Plus network. The MCS and HMI can be switched to serve the real rolling mill through a soft switching between aforementioned two signal interfaces of MCS to the real mill or the VMS. It is for this reason that the virtual reality technology becomes increasingly useful.

Multiple interstand fuzzy tension control

For 5 roughing rolling stands, there are 5 interstand zones appealing for tension control. Since the tension results equivalently from the speed mismatch of the upstream and downstream stand driving motors, it is possible to control (reduce or eliminate) the interstand tension by correcting the motor speed of either upstream or downstream stand.

Interstand tension affects the dynamics of both upstream and downstream stands. This makes the multiple interstand tension control a strongly coupled and highly ordered problem. The more the rolling stands, the more complex the control problem. Apparently, each interstand zone cannot be individually controlled well without a suitable decoupling strategy.

The rolling dynamics of each stand is highly nonlinear and difficult to model due to the complex metal deformation process. Intelligent fuzzy control is then a good candidate to deal with such a nonlinear and imprecise problem.

 

 

                                                                                

 

 

Reference RPM 1

+                                                      +

CC

+                                                       +

 

+                                                         +

+     Scheduled RPM 1                 +

Reference RPM 2

 

+

CC

Scheduled RPM 2

 

Reference RPM 5

+

+

 

+        Scheduled RPM 5

 

Stand 1

Stand 2

Stand 5

 

(FTC: Fuzzy tension controller; CC: Cascade controller; IC: Interval controller; ASR: Automatic speed regulator)