Volume 3, Issue 3, July 2013, Pages 693–700
V. Thamarai Selvi1 and Veluchamy Malathi2
1 Electrical and Electronics Engineering, Anna University Regional Centre Madurai, Madurai, Tamil Nadu, India
2 Electrical and Electronics Engineering, Anna University Regional Centre Madurai, Madurai, Tamil Nadu, India
Original language: English
Copyright © 2013 ISSR Journals. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
In Order to avoid mal operation of differential relay in transformer it is essential to distinguish between inrush and fault conditions. For accurate discrimination between inrush and fault current SVM technique is proposed. The merit of this method is demonstrated by simulation of different faults and switching conditions using MATLAB/SIMULINK. The inrush current values are obtained by varying the switching angle and the fault currents are obtained by varying the fault resistance. The Proposed method is tested on a 3000MVA, 230 kV Y-Y connected transformer by varying fault resistance, and switching angle. The performance of SVM is compared in terms of classification accuracy. The accuracy obtained using SVM is found to be more than other methods such as neural networks, ANFIS, etc. The results obtained with SVM are far better than other methods earlier used. SVM is preferred here over other methods because it is based on structural risk minimization whereas neural networks and ANFIS are empirical based. Moreover this method seems to be very effective for modern transformers with high harmonic contents and it requires less training. A SVM based protective field programmable gate array relay logic can be implemented further in future which will be verified against the simulation results.
Author Keywords: Inrush Current, Support Vector Machine, Fault current, Sequential minimization method, Radial basis function.
V. Thamarai Selvi1 and Veluchamy Malathi2
1 Electrical and Electronics Engineering, Anna University Regional Centre Madurai, Madurai, Tamil Nadu, India
2 Electrical and Electronics Engineering, Anna University Regional Centre Madurai, Madurai, Tamil Nadu, India
Original language: English
Copyright © 2013 ISSR Journals. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
In Order to avoid mal operation of differential relay in transformer it is essential to distinguish between inrush and fault conditions. For accurate discrimination between inrush and fault current SVM technique is proposed. The merit of this method is demonstrated by simulation of different faults and switching conditions using MATLAB/SIMULINK. The inrush current values are obtained by varying the switching angle and the fault currents are obtained by varying the fault resistance. The Proposed method is tested on a 3000MVA, 230 kV Y-Y connected transformer by varying fault resistance, and switching angle. The performance of SVM is compared in terms of classification accuracy. The accuracy obtained using SVM is found to be more than other methods such as neural networks, ANFIS, etc. The results obtained with SVM are far better than other methods earlier used. SVM is preferred here over other methods because it is based on structural risk minimization whereas neural networks and ANFIS are empirical based. Moreover this method seems to be very effective for modern transformers with high harmonic contents and it requires less training. A SVM based protective field programmable gate array relay logic can be implemented further in future which will be verified against the simulation results.
Author Keywords: Inrush Current, Support Vector Machine, Fault current, Sequential minimization method, Radial basis function.
How to Cite this Article
V. Thamarai Selvi and Veluchamy Malathi, “A classification approach using SVM to detect magnetic inrush in power transformers,” International Journal of Innovation and Applied Studies, vol. 3, no. 3, pp. 693–700, July 2013.
