Comparison of ICA and PCA methods for transformer internal fault studies

Traditionally discrete Fourier transform, wavelet techniques etc. are used for numerical transformer protection. In this study two different techniques are suggested for this purpose. The techniques so called dynamical principal component analysis (PCA) and independent component analysis (ICA) are investigated and compared to detect transformer internal faults. A three phase custom-built transformer is used for real time studies to examine the performance of the suggested techniques. Real time studies cover turn to turn faults and magnetizing inrush conditions. The suggested techniques are tested in MATLAB trade environment.

[1]  Guillaume Gelle,et al.  BLIND SOURCE SEPARATION: A TOOL FOR ROTATING MACHINE MONITORING BY VIBRATIONS ANALYSIS? , 2001 .

[2]  J. Faiz,et al.  A novel wavelet-based algorithm for discrimination of internal faults from magnetizing inrush currents in power transformers , 2006, IEEE Transactions on Power Delivery.

[3]  M. S. Sachdev,et al.  Online identification of magnetizing inrush and internal faults in three-phase transformers , 1992 .

[4]  S. Poyhonen,et al.  INDEPENDENT COMPONENT ANALYSIS OF VIBRATIONS FOR FAULT DIAGNOSIS OF AN INDUCTION MOTOR , 2003 .

[5]  Li Ping,et al.  Improved PCA with optimized sensor locations for process monitoring and fault diagnosis , 2000, Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187).

[6]  Raj Aggarwal,et al.  A wavelet transform based decision making logic method for discrimination between internal faults and inrush currents in power transformers , 2000 .

[7]  M. J. Fuente,et al.  Neural PCA based fault diagnosis , 2001, 2001 European Control Conference (ECC).

[8]  Venkat Venkatasubramanian,et al.  PCA-SDG based process monitoring and fault diagnosis , 1999 .

[9]  S.I. Shah,et al.  Dynamic process monitoring using multiscale PCA , 1999, Engineering Solutions for the Next Millennium. 1999 IEEE Canadian Conference on Electrical and Computer Engineering (Cat. No.99TH8411).

[10]  M. S. Sachdev,et al.  Design, implementation and testing of a microprocessor-based high-speed relay for detecting transformer winding faults , 1992 .

[11]  M. Suzuki,et al.  Digital protection method for power transformers based on an equivalent circuit composed of inverse inductance , 1988 .

[12]  P. Palmer-Buckle,et al.  Simulation of incipient transformer faults , 1998, 1998 Midwest Symposium on Circuits and Systems (Cat. No. 98CB36268).