Application of Breeder GA to Power System Controller Design

This paper presents the tuning of power system stabilizer (PSS) parameters using a relatively new evolution algorithm called Breeder Genetic Algorithms (BGAs). BGAs are based on the concept of "the survival of the fittest" typical to Genetic Algorithms (GAs). The main difference between GAs and BGAs is that the evolution of BGAs' population is based on artificial selection similar to the one used by human breeders. However, unlike GAs, the chromosomes in BGAs are always represented as sequences of real numbers rather than sequences of bits or integers. BGAs are particularly suitable to deal with continuous optimization parameters and are a very powerful and versatile optimization algorithm. The proposed BGA-PSS presented in this paper was tested over a wide range of operating conditions and its performance compared with both the Genetic Algorithm based PSS (GA-PSS) and the Conventional PSS (CPSS). Simulation results show that the performance of the BGA-PSS is better than that of the GA-PSS and the CPSS. However, both the BGA-PSS and the GA-PSS outperform the CPSS.