Robust control of a Sendzimir mill

In this paper a robust linear quadratic Gaussian (LQG) controller is designed for a multiple-input multiple-output industrial process. The plant is a Sendzimir mill, which is used in the steel industry to produce stainless steel strips. The design technique is to augment the plant with a set of integrators and apply guaranteed cost LQG theory to a corresponding set point error system. The advantage of this technique is that the integral action means the plant output is guaranteed to follow a setpoint for all admissible constant plant uncertainties. In other words, we design a robust tracking system for the mill