GENERALIZED PREDICTIVE CONTROL.

Abstract

In this work, the generalized minimum-variance (GMV) method was developed, which can stabilize systems with variable dead-time by the use of a sufficiently large control weighting. This method is a new member of the family of 'long-range predictive controllers'. The method is applicable to a plant which is both unstable and inverse-unstable (nonminimum-phase) and with an overparameterized model. Hence it can be used in cases where the standard GMV and pole-placement algorithms fail. It can also be used in areas such as robotics where future set-point changes are known.