EVALUATION OF SCOPE AND FEASIBILITY OF ARTIFICIAL INTELLIGENCE TECHNIQUES IN IMPROVING AUTOMATIC GENERATION CONTROL

Abstract

The operating point of the power system changes in a daily cycle due to the inherent nature of the changing load. This poses the difficulty in optimizing the conventional controller gains. Thus it may fail to provide the best dynamic response. To improve the performance of the Automatic Generation Control and to overcome the limitations of Conventional controller has necessitated the use of intelligent systems. Artificial Intelligence techniques like Fuzzy Logic, Particle Swarm Optimization and Genetic algorithms can be applied. A Fuzzy Logic Controller (FLC) is designed. It provides the control action in qualitative and symbolic form. The dynamic response is compared with the dynamic response obtained by the integral controller. Although the designed Fuzzy Logic controller in the two area hydro thermal for Automatic Generation Control system is available, that may not provide the better dynamic performance in all cases, so in order to improve the dynamic performance of the system a new Fuzzy Logic controller is designed by proper selection of the feedback gain and studying all the variable parameters effect, like Effect of different number of membership functions, rule base of the Fuzzy controller and speed regulation parameters on the dynamic response has been studied. The dynamic response is compared with the dynamic response obtained by the integral controller and the existing FLC controller. And still to improve the performance of the New Fuzzy logic Controller by tuning the parameters like membership function widths and scaling factors with the help of combined intelligence techniques PSO tuned Fuzzy Logic Controller and Genetic Algorithm tuned Fuzzy Logic Controller were proposed. The dynamic response is compared with the dynamic response obtained by the existing FLC controller and the new Fuzzy logic controller. The effect of unequal area capacity with respect to the dynamic response and robustness of the new fuzzy logic controller ,exixting fuzzy logic controller and the conventional controller were also studied.