A PMU-data-driven disruptive event classification in distribution systems

Abstract

With the proliferation of micro-phasor measurement units (μPMU) and PMUs in smart grids, time-synchronized high-resolution measurements can be obtained and used for numerous applications such as state estimation and event analysis. Disruptive events frequently occur in power grids and interrupt the normal operation of the system and may eventually cause the permanent failure of equipment. Therefore, establishing a data-driven event diagnostic framework to extract useful information such as the cause or location of events is of utmost importance. The disruptive events may not cause immediate and direct failure. However, they are a potential source for permanent equipment failure over time. Accurate disruptive event analysis is beneficial in terms of time, maintenance crew utilization, and future outages prevention. In this paper, a PMU data-driven framework is proposed to distinguish two disruptive events, i.e., malfunctioned capacitor bank switching and malfunctioned regulator on-load tap changer (OLTC) switching from two normal operating events i.e., the normal abrupt load change and the reconfiguration in distribution grids. The event classification is formulated using a neural network based algorithm, i.e., autoencoders along with softmax classifiers. The performance of the proposed framework is verified using the simulation of the events on the modified IEEE 123-bus distribution test system. The end results of this paper demonstrate the effectiveness of the proposed algorithm and satisfactory classification accuracies under several conditions such as different PMU reporting rates, different measurement noise levels, different number of PMUs, and boosting scenario.