Skip to Main Content
Estimation of Distribution Algorithms (EDAs) are a class of Evolutionary Algorithms in which the explicit representation of the population is replaced with a probability distribution where new candidate solutions are obtained by sampling this distribution. In lieu of crossover, the probability distribution vector is updated using candidate solutions that have the highest fitness. New sample solutions are generated from the updated probability distribution vector and another generation begins. This process is repeated until some convergence criteria are met. In this paper, one of the simplest version of EDAs known as Population-Based incremental Learning (PBIL) is considered. The standard PBIL is modified by adapting its learning rate according to the generation. The proposed Adaptive PBIL (APBIL) is used to optimally tune the parameters of a power system stabilizer (PSS). The effectiveness of the Adaptive PBIL is demonstrated by comparing the APBIL-PSS with the PBIL-PSS and the conventional PSS (CPSS). Simulation results show that both the PBIL-PSS and the APBIL-PSS provide more damping to the system than the CPSS. However, the APBIL-PSS gives a better performance than the standard PBIL-PSS.