An analytical framework for knowledge-based tuning of servo controllers

Abstract

This paper investigates the knowledge-based tuner of a two-level control structure in which a crisp servo controller that occupies the lower level is tuned by a fuzzy logic tuner at the upper level. First the knowledge-based tuner is expressed in the conventional fuzzy-logic formulation of a linguistic rule base to which the compositional rule of inference may be applied. Next the concepts of rule dissociation and fuzzy resolution are introduced, and resolution relationships are defined to build an analytical framework for the tuning problem. Stability of the overall control system is discussed in terms of the stability of the original system. Computational requirements for the tuning system are analyzed to show that a substantial reduction in computational effort can be achieved through this analytical framework. An example of tuning a PID controller of a nonminimum-phase plant with transport delay is presented to illustrate the application of the concepts that are developed in the paper.