The Swing-Blocking Methods for Digital Distance Protection Based on Wavelet Packet Transform and Support Vector Machine

Abstract

This paper presents a method for power swing and fault diagnosis of power system based on Wavelet Packet Transform(WPT) and Support Vector Machine (SVM) classifier. The method adopts Least Square Support Vector Machine (LS-SVM) classifier to identify the power swing and fault types. The power swing blocking elements are based on monitoring the rate of change of wavelet packet energy and wavelet packet entropy of voltage and current signal, the positive current and zero sequence component. The process of training the LS-SVM using a K-folded cross validation process for determining the values of parameter σ and parameter γ in RBF kernel parameters can minimize the classification error. The proposed method can successfully detect power swing and provide power swing blocking signal for accurate distance protection.

An overview of power swing detection methods in distance relays and the factors involved

“…This method can differentiate fault from PS and also has the ability to detect faults that are simultaneous with PS. In [77], a WT-based approach and SVM is proposed. This method can differentiate PS from faults and has the ability to detect faults that are simultaneous with PS.…”

Progress on power swing blocking schemes and the impact of renewable energy on power swing characteristics: A review

“…Kampeerawat et.al proposed another approach in [66], where Wavelet Packet Transform (WPT) and LS-SVM were used. The authors generated a double circuit transmission line model and simulated a power swing by inducing a three-phase fault.…”

“…As mentioned in [11,64,65], SVM has been known as a very potential tool in solving the classification problems. Being introduced by Vapnik [65] from the Statistical Learning Theory (SLT), it is found that SVM performs better than Artificial Neural Networks (ANN) because, SVM's essence follows on Structural Risk Minimization (SRM) while ANN is based on the Empirical Risk Minimization (ERM) [11,12,19,64,66]. The typical steps for SVM comprised of extracting the input features, and training the SVM classifier.…”

Wavelet entropy and its application for flicker contribution evaluation

“…The application of entropy in the power systems was started because the system under investigation will have different entropy values under different states. In [46][47][48][49][50][51][52][53][54], various methods for power system state estimation, fault detection, and power quality disturbance detection have been presented. It is also seen from these papers that the main objective is the detection and classification of power systems events by using the entropy approach.…”