Classification of Power Quality Disturbances in Emerging Power System Using Discrete Wavelet Transform and K-Nearest Neighbor

Abstract

This paper presents a wavelet and machine learning-based approach for the classification of power quality disturbances (PQDs) in emerging power systems. Renewable energy resources-based distributed generation (DG) is rapidly being used in the emerging power system to address the ever-increasing energy demand. PQDs are thought to be caused by the power electronic converters used in DG systems, DG operating conditions, and other common factors such as faults, switching activities, and non-linear loads. These PQDs must be detected and classified since they can create a variety of difficulties in end-user equipment. The proposed algorithm comprises the simulation of the emerging power system with a solar PV system, creating PQDs cases such as voltage sag, voltage swell, and voltage interruptions, capturing voltage signals which will be further processed using a discrete wavelet transform for feature extraction. The features extracted from DWT analysis are further used to develop the machine learning-based classifier for classification of PQDs. The proposed algorithm has been tested on a variety of PQDs. The simulation result shows that the proposed algorithm is efficient and it outperforms in the classification of PQDs.