KNN Based Approach for Transmission Line Outage Detection Using Synchrophasor Data

This paper explores the application of K-Nearest Neighbour (KNN) algorithm for detection of contingency due to line outage using the data from phasor measurement unit (PMU). In power system, contingency occurs due to outage of single or multiple lines. Detection of such contingency is needed in order to take corrective actions so that system stability is maintained. In this study, both the voltage phasors and current phasors are used to formulate the mathematical model. A new type of objective function is developed using the concept of Manhattan distance. The proposed method is tested on IEEE 5,14, and 57 bus networks. Detection rate of the suggested method is compared with some published methods and better performance is observed. The presented test results prove the effectiveness and viability of the proposed mechanism for detection of contingency.