Abstract: This study proposes an approach for classifying single stage Power Quality Disturbances (PQDs) that is based on the Hilbert Transform. These power quality disturbance signals are produced utilizing the MATLAB environment's simulation models and integral mathematical models of PQDs. Voltage sag, voltage swell, and voltage interruption are the power quality disturbances taken into account in this study. The Power Quality Index (PQI) curve is generated from these signals using the Hilbert transform, and features are then extracted from it. By changing the parameters, multiple instances of power quality disturbances are produced, and a dataset is produced. In order to train and test the Feed Forward Neural Network (FFNN) classifier, the dataset signals are transformed using the Hilbert transform to produce a feature vector. Calculating the efficiency of the proposed algorithm yields the effectiveness of the provided strategy. The single stage power quality disturbances are classified using a thresholding-based method as well.
Keywords: Power Quality Disturbances, Power Quality Index, Hilbert Transform, Feed Forward Neural Network