A gap metric based weighting method for multimodel predictive control of MIMO nonlinear systems

Abstract

A novel weighting method is proposed for multimodel predictive control of nonlinear systems with multiple scheduling variables (MIMO nonlinear systems), in which the gap metric is employed to formulate weighting functions for local controller combination. Compared to existent weighting functions, the proposed weighting method has two major advantages: firstly, there is only one tuning parameter, which makes it simpler. Secondly, the weights depend only on the scheduling vector and can be calculated off-line and stored in a look-up table. Therefore, the computational load can be reduced, especially for nonlinear systems with multiple scheduling variables. A MIMO CSTR system is studied to demonstrate the effectiveness of the proposed weighting method.