Identification of Combined Power Quality Disturbances in the Presence of Distributed Generations Using Variational Mode Decomposition and K-Nearest Neighbors Classifier

Identification of combined power quality disturbances in the modern power systems by considering the development of different types of loads and distribution generations has become increasingly important. the novelty of this paper comes from the accurate and fast identification of the combined power quality disturbances in the presence of different distributed generations and loads such as photovoltaic cell, wind turbine with doubly fed induction generators, diesel engines, electric arc furnace, dc machine, 6-pulse and 12-pulse rectifier loads. in this paper, the features are extracted using variational mode decomposition, just from voltage waveforms. to reduce the redundant data, dimension of features vector, and time, the relief-f method and correlation feature selection method are applied on the extracted features and these two methods are compared together. in this paper, the k-nearest neighbors classifier is used to classify the multiple power quality disturbances. to verify the effectiveness of the proposed method, different scenarios such as misfiring, variation of sun radiation and wind speed, entrance and exit of loads, capacitors and distributed generators, different fault at the grid in half-load to full-load were simulated. this method can be used as an added algorithm for smart metering in modern and smart power systems.