CLOSED LOOP TUNING OF DISCRETE PID REGULATORS USING GENETIC ALGORITHMS

In the present paper, an experimental method based on genetic algorithms for the closed loop parameter tuning of discrete PID regulators, is presented. Starting with the closed loop step response, the parameters of an approximated second order discrete model are identified in a first stage and after, the PID parameters which result in a significant response improvement, are determined. In both stages, a procedure based on genetic algorithms is used to find the optimal solutions. A simulated example of the proposed method is presented in this paper.