System Identification for Robust Control of Processes with System Changes

This study aims at integrating system identification for robust control with process control. We consider linear systems where the parameters normally vary slowly, but which may exhibit sudden and large changes. One example of such a system is an extruder where the parameters vary slowly e.g. due to temperature changes, and where the die sometimes is changed, leading to sudden and large changes. Due to a large number of possible dies, it is infeasible to develop models for all possible system configurations. It is therefore necessary to use system identification to develop models for the various configurations.