A Framework Based on Machine Learning for Analytics of Voltage Quality Disturbances

This paper proposes a machine-learning-based framework for voltage quality analytics, where the space phasor model (SPM) of three-phase voltages before, during, and after event is applied as input data. The proceeds along with three main steps: (a) extraction, (b) characterization, (c) additional information extraction. During first step, it utilizes Gaussian-based anomaly detection (GAD) technique to extract data from recording. Principal component analysis (PCA) adopted during second shown that principal components correspond semi-minor semi-major axis ellipse formed by SPM. third these characteristics are interpreted about underlying cause event. performance was verified through experiments conducted on datasets containing synthetic measured power events. results show combination axis, direction major forms sufficient base characterize, classify, eventually recorded