An LMI approach to IMC-based robust tunable control

This paper presents a linear matrix inequalities (LMI) approach to robust tunable control. This controller design technique provides a new on-line tuning strategy for industrial process control systems. The tuning strategy is based on the performance robustness bounds of the system and knowledge of the plant uncertainty's weighting function, which may change with time. The internal model control structure of the controller is adopted together with additive plant uncertainty. The design and tuning of the robust tunable controller for an SISO system relies on an approximate solution to the two-disc optimization problem, that is solved over the class of discrete-time finite impulse response filters via the solution of an LMI problem. A numerical example is given to illustrate the technique.