Wide Area Measurement System based Frequency Data Mining for Event Detection in Power System

The wide-area measurement system or WAMS has been commissioned at various substations. Phasor Measurement Units (PMUs) have been positioned at various 400kV substations and several Phasor Data Concentrators (PDCs) have been positioned at respective Regional Load Dispatch Centres (RLDCs). The volume of data excerpted is hefty. More often than not, the power system is in an ambient condition; ostensibly there are constantly a few occasions for every day, happening in the framework which may influence the strength and stability of our power system. Frequency is a ubiquitous signal in the power system. Which retorts to each significant perturbations like power swings, short circuit faults etc. This report introduces a strategy to consequently distinguish such disturbances while mining data from WAMS. The PMUs are able to compute the time-stamped value of frequency, rate of change of frequency, current and voltage in phasor format. Frequency measurements gathered from different locations are being brought to the algorithm of automatic event detection. The new technique has been enumerated to automatically recognize an event by utilizing frequency data acquired from only one location. L-1 Trend filter and Extended L-1 Trend filter is used to automatically detect the event and instance of the start of the event respectively. Two case studies and its outcomes for event detection has contrived here.