Identification of interval models for a class of uncertain systems via linear programming

The interval models of uncertain plants are frequently used in the field of robust control. In this paper, a novel interval model identification method based on linear programming is proposed. By certain prepossessing on the measured data of the plant, the interval model identification problem is transformed into several linear programming problems, which can be easily solved through various algorithms. We have proved that the resultant interval model has a known error bound, provided that certain conditions on the measured data are satisfied. To demonstrate its effectiveness, the proposed method is applied to identify the interval model of a servo system with variant load and inertia.