Fault Detection and Classification using Wavelet and ANN in DFIG and TCSC Connected Transmission Line

TL;DR

This paper introduces a Wavelet-ANN based fault detection and classification method for DFIG and TCSC connected transmission lines, demonstrating robustness under variable conditions like wind speed and impedance faults.

Contribution

It proposes a novel Wavelet-ANN approach that improves fault detection accuracy in variable conditions, outperforming existing methods.

Findings

Robust fault detection under varying wind speeds and impedance faults

Effective classification of fault types using Wavelet-ANN

Enhanced performance compared to state-of-the-art methods

Abstract

This paper presents fault detection and classification using Wavelet and ANN based methods in a DFIG-based series compensated system. The state-of-the art methods include Wavelet transform, Fourier transform, and Wavelet-neuro fuzzy methods-based system for fault detection and classification. However, the accuracy of these state-of-the-art methods diminishes during variable conditions such as changes in wind speed, high impedance faults, and the changes in the series compensation level. Specifically, in Wavelet transform based methods, the threshold values need to be adapted based on the variable field conditions. To solve this problem, this paper has proposed a Wavelet-ANN based fault detection method where Wavelet is used as an identifier and ANN is used as a classifier for detecting various fault cases. This methodology is also effective under SSR condition. The proposed methodology is evaluated on various fault and non-fault cases generated on an IEEE first benchmark model under varying compensation levels from 20% to 55%, impedance faults, and wind velocity from 6m/sec to 10m/sec using MATLAB/Simulink, OPALRT(OP4510) manufactured real-time digital simulator environment, Arduino board I/O ports communicating with external PC in which ANN model dumped, using Arduino support package of MATLAB. The preliminary results are compared with the state-of-the-art fault detection method, where the proposed method shows robust performance under varying field conditions.