NEURAL NETWORK BASED ADAPTIVE AUTORECLOSURE TECHNIQUE FOR LONG-DISTANCE COMPENSATED TRANSMISSION SYSTEMS

ABSTRACT This paper discusses the application of neural networks to adaptive autoreclosure technique for long-distance compensated systems. Firstly, the detailed simulation studies on a long-distance 500kV compensated systems are presented. From the simulation studies, some salient features of the faulted responses are revealed and summarized. Based on the findings of investigations, a neural network based adaptive autoreclosure scheme is developed for compensated systems. Special emphasis is placed on the selection of an appropriate neural network topology and its training