Adaptive Neuro-Fuzzy Control Approach Based on Particle Swarm Optimization INTRODUCTION Particle Swarm Optimization (PSO) has been an increasingly hot topic in the area of computational intelligence. PSO is an optimization algorithm that falls under the soft computing umbrella that covers genetic and evolutionary computing algorithms as well. As such, it lends itself as be- ing applicable to a wide variety of optimization problems (Esmin & Torres, 2004;, Parsopoulos & Skokos, 2005; Helwig & Haubelt, 2005; Conradie & Miikkulainen, 2002). PSO is a population-based algorithm that exploits a population of individuals, to search promising regions of the function space. In this context, the population is called swarm and the individuals are called particles. Each particle moves with an adaptable velocity within the search space, and retains in its memory the best position it ever encountered. In the global variant Power systems are modeled as large scale non-linear highly structured systems. The high complexity and nonlinearity of power systems have been created a great deal of challenge to power system control engineers for decades. One of the most important problems in the electric power systems is the damping of low-frequency oscillation (dynamic stability). Such oscillations may occur between the electrical and mechanical systems or between large inertia's in the mechani- cal system. These oscillations are usually initiated by small disturbances such as small changes in the load levels or generator loading. If the distur- bance is large (transient stability), the oscillations may be sustained for minutes and grow to cause system separation if no adequate damping at the system oscillating frequency is available (Ahmed, 2000). Therefore, a major effort has to be made to improve power system stabilizers (PSSs) per- formance and characteristics. PSSs are usually