Control configuration selection for multivariable switched dynamical systems and processes

The decentralized and partially decentralized control strategies have gained in a lot of popularity in industrial practices, and they are becoming even more in demand with the increasing complexity of large-scale processes and plants. A decentralized or partially decentralized control strategy cannot be successful unless proper input-output pairs are selected for the controller synthesis. Selecting proper control structure is therefore very important in practice. In this paper the issue of selecting proper control structures will be investigated for multivariable switched systems and processes. To the best of our knowledge, the reported results in this context, do not support switched systems and processes. However, in a lot of applications the underlying models for the systems and processes are of hybrid and switched nature. It is therefore necessary to define a measure which is capable of quantifying the interaction between different input-output pairs for such systems. The main contribution of this paper is to propose such measure based on the concept of gramians and to describe how to use this measure to select proper configuration for controllers for switched systems and processes. A numerical example explains the results in more details.