Incorporation of a generalized TSK model in nonlinear model predictive control

Three innovations are demonstrated as effective for nonlinear horizon predictive control. First, in this single-input-multiple-output (SIMO) application, a recently reported generalized Takagi-Sugeno-Kang (GTSK) model is used to predict the controlled variable. Second, a novel optimization technique, Leapfrogging, is used to solve for the horizon of future manipulated variable moves. Third, the “sawtooth” pattern is used as the input to generate the model. The demonstration is subject to both soft and hard constraints - soft on both the controlled and auxiliary variable, and hard on both the limits and rate of change of the manipulated variable.