Cascade Generalized Predictive Control—Applications in power plant control

Abstract The Generalized Predictive Controller in transfer function representation is proposed for the cascade control task. The recommended cascade GPC (CGPC) applies one predictor and one cost function that results in several advantageous features: The disturbance regulations of the inner and the outer loops can be totally decoupled; The inner disturbance regulation is well damped, the typical overshoot of the traditional cascade control structure is avoided; The robustness properties of the inner and the outer loops can be designed separately; The anti-windup properties of the CGPC are exactly as perfect as in the case of the simple SISO GPC. The typical problem of the saturation in the inner loop, resulting in modeling error for the outer loop, is prevented. The CGPC was applied as the oxygen controller of a pilot fluidized bed boiler. The investigation is based on simulation experiments and on experiments on a pilot scale boiler. In another simulation experiment, the CGPC was applied as the temperature controller of at a steam superheater stage. The results of the experiments well illustrated the power of the proposed cascade control algorithm