Detection of Power System Events Using Bagged Tree Classifier

The detection of events is crucial for robust operation and fast restoration of power system. This paper focuses on the identification of power system events, using an ensemble technique, i.e., bagged tree classifier. Five possible events have been considered in this work, which includes line trip, load trip, generator trip, fault, and transformer trip. The various events have been created in IEEE 9-bus system and data have been recorded through phasor measurement units (PMUs). The consideration of all five power system events assures robustness of the proposed method. The IEEE 9-bus system has been modelled and simulated in MATLAB/Simulink platform. The proposed approach promises excellent detection accuracy.