GPC USING A DELTA-DOMAIN EMULATOR-BASED APPROACH

This paper describes new approaches to generalized predictive control formulated in the delta (delta) domain. A new delta-domain version of the continuous-time emulator-based predictor is presented. It is shown to contain the optimal discrete-time predictor based on incomplete in formation as a special case. Usually, a good estimate is obtained in a much longer range of samples than obtained by the optimal predictor o f the same complexity. This is particularly advantageous at fast sampl ing rates where a 'conventional' predictor is bound to become very com putationally demanding. Two controllers are considered: one having a w ell-defined limit as the sampling period tends to zero, the other bein g a close approximation to the conventional discrete-time GPC. Both al gorithms are discrete in nature and well-suited for adaptive control. The fact that S-domain models are used does not introduce an approxima tion since such models can be obtained by exact sampling of continuous -time models.