Rolling Mill - System Modelling and Classical Control Virtual Laboratory

Summary

Rolling Mills are of major technological significance in modern society. The aims of rolling mills are to reduce the thickness and change the metallurgical properties of a metal strip as it is passed through a pair of driven parallel cylindrical rolls.

Rolled sheet metal is an important material in modern manufacturing, rolled products are found everywhere in our daily lives.

Of particular relevance to this virtual laboratory is the problem of automatic thickness control in rolling mills. This virtual laboratory will successively investigate a number of control architectures aimed at dealing with the issues raised in this application.

Figure 6.1: Screenshot of Program
Figure 6.1: Screenshot of Program

The Physical Apparatus

Rolling Mills are of major technological significance in modern society. Perhaps surprisingly, they are very high technology pieces of equipment with strip speeds over 100 km/hour and tolerances of 0.1%. Essentially, the aim of rolling is to reduce the thickness and change the metallurgical properties of metal strip as it is passed through a pair of driven parallel cylindrical rolls.

The high external roll forces cause a permanent plastic deformation of the strip, decreasing it’s thickness and producing a corresponding increase in length. The strip may be either hot or cold rolled. For example steel plate (typically > 1.5mm) is hot rolled and steel sheet (typically < 0.1mm) or aluminium foil are cold rolled.

Rolled sheet metal is an important material in modern manufacturing, rolled products are found everywhere in our daily lives. Consider rolled aluminium products:

  • Foil less than 0.2mm thick is used in packaging such as foil containers and wrapping, in electrical applications, building insulation and in printing industries.
  • 0.2mm to 6mm thick has a wide variety of uses in construction industries such as sidings and roofing. Sheets also used in transport for automotive body panels, airframes and boat hulls.
  • Plate over 6mm thick airframes, military vehicles, construction components for buildings and bridges.

Rolling Mill
Figure 6.2: Rolling Mill - Photo courtesy of BlueScope Steel

In the year 2000 alone, 3.44 million tonnes of aluminium rolled products were consumed in western Europe (European Aluminium Association’s web site http://www.eaa.net/). Rolling mills have been in use since the fifteenth century, however increasing performance demands have recently lead to greater sophistication in rolling mill technology.

Forces can be of the order of 2,000 tonnes, strip speeds up to 120 km/hour and (in the aluminium industry) thickness as low at 0.05mm and tolerances of 5 micrometers. (This represents a tolerance of 0.1% which is “high tech” by any standard.) The mechanical systems, materials, sensors, actuators and control systems required to achieve this have undergone substantial improvement over time.

Of particular relevance to this virtual laboratory is the problem of automatic thickness control in rolling mills; a problem that has been the subject of much investigation over the last five decades. Through the development of new sensors and actuators combined with the appropriate application of new control architectures, the fundamental performance limits of earlier architectures have been overcome and tolerances reduced by 2 orders of magnitude (from 10% to 0.1%).

In keeping with the evolutionary improvement in technology this virtual laboratory will successively investigate a number of control architectures aimed at dealing with the issues raised in this application.

Prerequisites

This virtual laboratory is suitable for use in a first course in control, students should be familiar with:

  • Laplace Transforms
  • Transfer Functions
  • Time Delays
  • PID Controllers
  • Smith Predictors

Learning Objectives

The objectives of this laboratory are:

  • To introduce the student to the set of problems associated with rolling mill thickness control
  • To gain an appreciation of simple control system design tools including Classical PID design, and Time delay compensation via a Smith Predictor
  • To gain insight into fundamental performance limitations associated with various control system architectures applied to this system