Power signal classification using dynamic wavelet network

A comprehensive overview on signal processing and artificial intelligence techniques applications in classification of power quality disturbances

“…A combination of fully informed particle swarm (FIPS) and adaptive PNN approach known as PNN based feature selection (PFS) was proposed in [136] to identify the PQ disturbances where the statistical features were extracted by ST and time-time (TT) transform (extension of ST) for training the PNN classifier. An integrated approach of wavelet layer and adaptive PNN known as Dynamic Wavelet Network (DWN) was proposed in [137] particularly suitable for the classification of PQ signals in a dynamic environment with time varying nonstationary PQ signals. The DWN is the combination of the two sub-networks consisting of input DWT layer and adaptive PNN layers.…”

Classification of power quality events – A review

“…Basically, non-stationary signals (having multiple frequencies ranging from 300 to 1000 Hz such as oscillatory-transients, voltage spikes, multiple voltage notches due to solid-state converter switching and harmonics) characterized by wide range of frequency spectrum with transient and sub-harmonic components are difficult to analyze as in [24]. These disturbances can be monitored as in [25] and classified on the basis of time-variant statistical characteristics of the voltage and current waveforms as in [26][27][28] and they could be sinusoidal or non-sinusoidal as in [14]. For only non-stationary PQ events, dominating frequency components have been used as features for recognition of events in [29].…”

An optimum ECG denoising with wavelet neural network

“…Conjugate gradient method isconsidered as somewhat in between steepest descent Newton's method having positive features of both of them. Due to faster convergence of conjugate gradient method is used as the network training function [4,7,6]. Now a day Neural Networks are widely used in biomedical applications.…”