Znalostní adaptace regulátoru s implementací v programovatelných automatech

Abstract

The thesis is focused on the research and development in knowledge adaptation of conventional PID controllers. The proposal of the methodology for the knowledge adaptation of controllers with verification of the implemented algorithms on simulated and real systems is the main purpose of the thesis. General knowledge model for the adaptation with verification during the control experiments is defined. Identification algorithms for determination of controlled system model parameters, effective control of the identification experiments and algorithms, algorithms for adaptation of the conventional PID controllers on the basis of the identified system and methods for determining of quality of regulation are the core of the knowledge adaptation model. The functionality of the proposed methodology has been verified in practical regulation experiments.