This paper demonstrates the application of artificial neural network with a flexible structure, as a powerful design tool, in automatic load-frequency control design in a restructured power system. A typical power system in a competitive distributed control environment with an open access organizational structure is considered. A new neural network based control strategy is used to maintain the system reliability and eliminates the frequency error. It is shown that in a deregulated environment, the conventional control methodologies are incapable of obtaining good dynamical performance. In order to achievement of all specified control objectives, dynamic neurons with a wide range of variation, and an adequate rate of flexibility are used within the structure of proposed neural network. Some nonlinear simulations are used to illustrate the design methodology.
