EVENT ANALYSIS OF DISTRIBUTION-LEVEL POINT-ON-WAVE DATA

Abstract

The connection of distributed generation to distribution networks has necessitated that the monitoring system be revaluated as dynamics of the system continue to change and increase in speed. In an advanced monitoring system, high resolution measurements are the main component. This data can be in the form of time-sequential samples of voltage and current waveforms referred to as point-on-wave data. In this work, a dataset of point-on-wave data from three locations is time-synchronized and analyzed for events. The sampling rate of this data is 2000 Hz. The metric used to screen for faults and transients is the magnitude of current. Frequency and voltage events are detected from the variation in frequency over each minute. In this work frequency is calculated from the phase of the voltage waveforms. These events are categorized as either originating from the transmission or distribution network based on the magnitude and timing of the responses as recorded by the measurement devices. In addition, the modal content and damping of low frequency oscillations that are detected in the data are examined using a nonlinear least-squares optimization-based approach that has been modified for power systems. The results corresponding to different locations and oscillations are compared to determine characteristics of the system’s behavior in response to events that trigger these oscillations in the power system.