Detection of High Impedance Fault in Power System Using MWT and ANFIS

N. Narasimhulu, Dr.D.V. Ashok Kumar, Dr.M. Vijay Kumar

In this paper, the detection as well as the classification of high impedance faults (HIF) is analyzed using wavelet and artificial intelligence (AI) technique. Here, the Multi Wavelet Transform (MWT) and the Adaptive Neuro Fuzzy Inference System (ANFIS) are employed for detecting the HIF in power systems. The MWT is an extension of wavelet transform, which is used to extract the feature of the signal. ANFIS is one of the Artificial Intelligence (AI) techniques, which allows the classification of the signal in terms of its features. Initially, the normal signals are analyzed. Later, the system‟s behavior after the fault occurrence is analyzed and the faulty signals can be seen as distorted waveforms. This type of distorted waveforms is composed of various frequency components and required to represent in time-frequency to carry out the fault analysis. For achieving this representation of line signal, MWT is presented. It extracts the faulty features, which are then forwarded to ANFIS for classifying the type of fault that has occurred in the power systems. The proposed methodology is carried in the MATLAB/Simulink platform and the performance of fault detection process is evaluated. The evaluated results are compared with Discrete Wavelet Transform (DWT) - ANFIS, MWT – Neural Network (NN) and DWT – Radial Basis Function Neural Network (RBFNN).The comparison of performances reveals that the proposed technique is better than the other techniques in terms of statistical measures.

Volume 12 | 05-Special Issue

Pages: 1490-1506

DOI: 10.5373/JARDCS/V12SP5/20201911