Study of turbine-generator torsional oscillations with particular reference to adaptive autoreclosure

Adaptive autoreclosure techniques can offer many advantages over conventional schemes in improving power system stability and reliability. Previously (1993), the authors developed a novel autoreclosure scheme by using a neural network approach. The results show that the proposed scheme can distinguish between permanent and transient faults, and can precisely define the arc extinction time in the case of transient faults. In this paper, studies of turbine-generator torsional oscillations are conducted. The outcome of the studies offers a better understanding in the development and applications of adaptive autoreclosure.<>