Secured zone-3 protection during power swing and voltage instability: an online approach

Maloperation of distance relay under stressed conditions like power swing and voltage instability has been reason of various power system blackouts. Thus, it has become very important to make the decision making of the distance relay to be intelligent enough so that it is capable of differentiating between fault and other stressed conditions. There is no shortage of computational intelligent techniques suggested in the literature for this purpose. However, with the growing complexity of the power systems, there have been cases when various system parameters, system topology and so on suddenly change. This study suggests an online intelligent technique: online sequential extreme learning machine that gives accurate result on such circumstances owing to its features of being an online tool. The suggested technique based on wide area measurements consists of two levels of classifiers that segregate stressed conditions like power swing and voltage instability from fault. Performance of the online intelligent technique has been compared with other intelligent techniques defined in the literature. Moreover, by using synchronised phasor measurements, shortcomings of the conventional relay technology have been overcome.