Mobile Information Systems, cilt.0, sa.0, ss.1-11, 2021 (SCI-Expanded)
Power quality disturbance (PQD) is essential for devices consuming electricity and meeting today’s energy trends. *is study
contains an effective artificial intelligence (AI) framework for analyzing single or composite defects in power quality. A con-
volutional neural network (CNN) architecture, which has an output powered by a gated recurrent unit (GRU), is designed for this
purpose. *e proposed framework first obtains a matrix using a short-time Fourier transform (STFT) of PQD signals. *is matrix
contains the representation of the signal in the time and frequency domains, suitable for CNN input. Features are automatically
extracted from these matrices using the proposed CNN architecture without preprocessing. *ese features are classified using the
GRU. *e performance of the proposed framework is tested using a dataset containing a total of seven single and composite
defects. *e amount of noise in these examples varies between 20 and 50 dB. *e performance of the proposed method is higher
than current state-of-the-art methods. *e proposed method obtained 98.44% ACC, 98.45% SEN, 99.74% SPE, 98.45% PRE,
98.45% F1-score, 98.19% MCC, and 93.64% kappa metric. A novel power quality disturbance (PQD) system has been proposed,
and its application has been represented in our study. *e proposed system could be used in the industry and factory.