Initial investigation of data mining applications in event classification and location identification using simulated data from MinniWECC

Phasor measurement units can provide high resolution, synchronized measurements of a power system. These measurements result in massive amounts of data which is tall over time (30 samples per second) and wide over many channels. Data mining techniques are promising tools for quick and efficient examination of power system events with respect to multidimensional correlated data. In this paper, we discuss the implementation and performance of various data mining techniques for classifying common event types and the associated event location. Results are presented using test data from the MinniWECC model of the Western Electricity Coordinating Council (WECC) which is designed to maintain inter-area modal properties. Results show promising performance of these techniques for identifying both events (classification rates exceeding 99%) and event locations (classification rates exceeding 80%). This paper also discusses strategies and challenges for implementing such data mining techniques to real world PMU data.