Recognition of Voltage Sag Causes using Vector Quantization based Orthogonal Wavelet

As per mitigation aspects, recognition of the events which cause various power quality disturbances holds more significance as compared to recognition of those power quality disturbances. So, this paper proposes an approach for recognition of events which causes voltage sag. This work designs new orthogonal signal-adapted wavelet basis using vector quantization for more efficacious multiresolution analysis of input signals. From filterbank of designed wavelets, most optimal wavelet basis for a particular input signal is chosen by adopting the criterion of maximum energy-to-Shannon-entropy ratio. Multiresolution analysis of the input signal using selected wavelet basis produces more distinguishing and informative feature set to be classified by naive Bayes classifier. The proposed method has achieved good classification accuracy even in the presence of noise which proves its robustness for real-time applications.