Recognition of power quality events by using multiwavelet-based neural networks

Classification of power quality events – A review

“…WT has been proved an effective tool for detecting and classifying the PQ events. Paper [58] utilizes an enhanced resolving capability of multiwavelet to recognize power system disturbances. The papers by Behera et al [97], Panigrahi et al [68] and Uyar et al [140] implemented S-transform because ST has an advantage in that it provides multiresolution analysis while retaining the absolute phase of each frequency.…”

“…In particular, whereas WT has an associated one scaling and wavelet function, MWT has two or more scaling and wavelet functions. In [58], authors developed a framework for the recognition of PQ disturbances using MWT and neural networks. The computation cost is quite higher than WT because it contains more than one wavelet and scaling function.…”

Real Time PC Implementation of Power Quality Monitoring System Based on Multiharmonic Least-Squares Fitting

“…Over the past years, several methods for detection and classification of power quality events in a power system were presented [5][6][7][8][9][10][11][12][13][14][15][16][17]. The wavelet transform is one of the most often employed signal-processing techniques used for power quality detection algorithms [5][6][7][8][9][10][11].…”

“…The wavelet transform is one of the most often employed signal-processing techniques used for power quality detection algorithms [5][6][7][8][9][10][11]. It can successfully detect high frequency events (transients) but in the case of slow disturbances, it performs poorly especially when the voltage variations are not sudden, but gradual.…”

Wavelet transform and multi-class relevance vector machines based recognition and classification of power quality disturbances

“…Some examples are fast Fourier transform method [3], fractal-based method [4], S-transform method [5], time-frequency ambiguity plane method [6], short time power and correlation transform method [7], wavelet transform method [8], Hilbert transform method [9], and Chirp-Z transform (CZT) method [10]. Neural networks (NNs) [11,12] with different structures are traditionally used as classifier but recently probabilistic neural network (PNN) [2,13] and support vector machines (SVMs) [14][15][16] are introduced as new learning machines which are more effective. Hidden Markov model [17], rule-based expert system and fuzzy logic have also been employed at the decision-making step in the process of classifying PQ disturbance types [18,19].…”