An algorithmic method for control structure selection based on the RGA and RIA interaction measures

A systematic framework for the generation of the most promising control structures for large dimensional systems is proposed in this paper. The new algorithm is based on the main properties of the RGA (Relative Gain Array) and RIA (Relative Interaction Array) matrices and the concepts of interaction, integrity and stability. It is shown that the minimization of the overall interaction in terms of the RIA matrix for multiinput-multioutput systems under several stability and structural constraints can be formulated as a mixed integer linear programming (MILP) problem. In order to demonstrate the usefulness of the proposed algorithm as a rigorous and systematic solution to the problem of the automatic generation of promising control structures, two large scale industrial problems are considered.