An Embedded-Based Distributed Private Cloud: Power Quality Event Classification

This study designs and implements an embedded system-based distributed private cloud (EBDPC), which is a real-time analyzing device used in a wireless sensor network (WSN) monitoring system application. The proposed system consists of several low-cost embedded systems for the detection, classification, and measurement of six types of power quality events, including impulse, voltage sag, voltage swell, interruption, harmonics, and flicker. Fast Fourier Transform (FFT) methods are used for automatic signal classification and characterization, and real-time notification for disturbance detection. A light-weighted load balance algorithm is also included in the proposed system for handling a great amount of data collected from electric sensors. The system has been successfully applied to detecting, classifying, and measuring disturbances in a power quality monitoring WSN.