Data-mining model based adaptive protection scheme to enhance distance relay performance during power swing

The paper presents a data-mining model based adaptive protection scheme enhancing distance relay performance during power swing for both compensated and uncompensated power transmission networks. In the power transmission network, the distance relays are sensitive to certain system event such as power swings, which drive the apparent impedance trajectories into the protection zones of the distance relay (zone-3) causing mal-operation of the distance relay, leading to subsequent blackouts. Further, three-phase balanced symmetrical fault detection during power swing is one of the serious concerns for the distance relay operation. This paper proposed a new adaptive protection scheme method based on data-mining models such as DT (decision tree) and RF (random forests) for providing supervisory control to the operation of the conventional distance relays. The proposed scheme is able to distinguish power swings and faults during power swing including fault zone identification for series compensated power transmission network during stress condition like power swing. The proposed scheme has been validated on a 39-bus New England system which is developed on Dig-Silent power factory commercial software (PF4C) platform and the performance indicate that the proposed scheme can reliably enhance the distance relay operation during power swing.