Open AccessDiscriminant Analysis and Hilbert Huang Based Power Quality Assessment
Author(s) -
Stuti Shukla Datta,
Namrata Dhanda
Publication year - 2019
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.a9635.109119
Subject(s) - hilbert–huang transform , hilbert transform , wavelet transform , maxima and minima , pattern recognition (psychology) , computer science , mathematics , signal processing , artificial intelligence , s transform , algorithm , harmonic wavelet transform , fourier transform , wavelet , hilbert spectral analysis , discrete wavelet transform , spectral density , digital signal processing , statistics , mathematical analysis , computer hardware , white noise
This work deals with Hilbert Huang transform and discriminant analysis based assessment of power signals. Hilbert Huang transform is a combination of Empirical mode decomposition (EMD) and Hilbert Transform. EMD is a data assisted processing technique that works on the time scale difference between local extremas (maxima and minima points of a signal). Unlike Fourier Transform, Wavelet Transform and Stockwell Transform, EMD does not employ any basis function or a window function and highly depends on the data of the signal. Power system is a highly vulnerable system subjected to several technical constraints and hence deviation of power signals from their normal level is inevitable. Thus, in order to study the reasons that cause the deviation of normal values, signal processing technique based on EMD is applied to power signals which are obtained by simulating various power scenarios in MATLAB Simulink platform. Decomposed components are then transformed in the frequency domain using Hilbert Transform. Hilbert transform helps in the extraction of features of the signal in consideration. These features are then subjected to discriminant analysis based classifier to identify the class of the raw input. Efficiency of the methodology is evaluated and results obtained are highly promising.