Weighting goals and constraints in fuzzy predictive control

Model predictive control (MPC) is a well-known control technique, which has been applied to complex and nonlinear processes. Fuzzy predictive control incorporates fuzzy goals and constraints in MPC, by combining predictive control with fuzzy decision making. In this paper, we propose the integration of weighted criteria in fuzzy predictive control, where the decision-maker can specify the preference for different goals and constraints by using weight factors for each criterion. A new heuristic is proposed to select suitable weight factors that satisfy the overall control objective. In this context, a method to extend the t-norms to the weighted case is also discussed. The weighted approach is validated using a multivariable process: the simulation of a gantry crane system, which shows clear improvements in the control performance when using the weighted fuzzy predictive control approach.