REPLACEMENT OF CONVENTIONAL P55 WITH A GENETIC TUNED PID CONTROLLER BASED ON DIFFERENT SEARCH CRITERION

This paper proposes a genetically tuned PID controllers for power system models with uncertain
parameters to replace the conventional power system stabilizer. This may overcome the problems
arising due to the fact that PSS is only effective for a linear power system model with certain
parameters and disturbances with relatively limited small range around the nominal operating
condition. The first step in the design procedure is to find out appropriate PID parameters which
are essential to assess and initiate the genetic search within a confident evolution environment.
The industrially recognized Ziegler-Nicholes methodology will be employed for this purpose.
Secondly, different search criteria such as Integral of Time multiplied by Absolute Error (ITAE),
Integral of Absolute Magnitude ofthe Error (IAE), Integral ofthe Square of the Error (ISE), and
Mean ofthe Square ofthe Error (MSE) are implemented to ensure the robustness ofthe proposed
controller. Several experiments will be undertaken to evaluate which of these four performance
criteria produce the best results when used in conjunction with a Genetic Algorithm (GA). The
Results of implementing the proposed GA-tuned PID controller show that the most satisfactory
response, will be obtained if a GA with MSE or ISE criterion is selected to tune the PID controller.