Selection of control structure for distributed model predictive control in the presence of model errors

Abstract

This paper presents a new methodology for selecting control structure in the context of distributed model predictive control. An index was developed to quantify the performance of distributed MPC strategies in the presence of model errors. This index was used for two purposes: to solve the decomposition problem whereby the process is decomposed into parts and to compare distributed MPC strategies with different degrees of coordination. Then, a multi-objective Mixed Integer Nonlinear Programming MINLP formulation is proposed to achieve an optimal trade-off between performance and structure simplicity. Four examples are considered to illustrate the methodology. The simulation results are consistent with the conclusions obtained from the analysis. The proposed methodology can be applied at the design stage to select the best control configuration in the presence of model errors.