Multi-model adaptive predictive functional control and its application

Predictive functional control (PFC) is usually used because of its strong robustness that can overcome the plant´s time varying in a not wide range, but if the plant´s property varies in a very wide range, ideal-regulating results could not be obtained while fixed model is used. In view of slow time-varying system, an adaptive multi-model PFC method using first-order plus pure time-delay simplified model is presented, adaptive parameter identification algorithm is convergent, the method is simple and easy to be realized. The strategy is applied to a main steam temperature system of a coal-fired power plant; computer simulation results show that the method is effective; good regulating performance is obtained while controlled plant varies.