Wavelet-based partial discharge image denoising

An application of wavelet-based denoising to phase-resolved partial discharge images is presented. The basic principles of wavelet denoising analysis, with a special focus on image decomposition, as well as examples of hard- and soft denoising thresholding are reported. For the purposes of decomposition, the Deabuchies wavelet and wavelet packets at different levels were applied. Simulations are discussed and the results obtained during online measurements on a 6 kV/200 kW motor are presented. The method described is especially suited to cases in which an external additive noise uncorrelated with a partial discharge (PD) signal is present during acquisition, for example, in cables, transformers, rotating machines and gas-insulated switchgears. The fundamental issue in image recovery using wavelet denoising seems to be the choice of the threshold value and the type of the wavelet. Proper preprocessing is crucial prior to pattern recognition on the basis of a correlation with predefined PD forms. In addition, wavelet decomposition could be treated as lossy image compression in applications such as image internet transfer to/from external databases, in which only wavelet coefficients could be sent discarding the ones below a certain threshold level. The method presented can be applied during PD acquisition, for example, in high voltage cables, transformers, rotating machines and gas-insulated switchgears. The wavelet denoising processing will definitely find future applications in PD analysers, besides the boxcar accumulation method and spatial or FFT-based digital filtering

[1]  Xiandong Ma,et al.  Interpretation of wavelet analysis and its application in partial discharge detection , 2002 .

[2]  A. Krivda,et al.  Automated recognition of partial discharges , 1995 .

[3]  T. Endoh,et al.  Wavelet analysis of tree developments , 1995, Proceedings of 1995 IEEE 5th International Conference on Conduction and Breakdown in Solid Dielectrics.

[4]  L. Satish,et al.  Wavelet-based denoising of partial discharge signals buried in excessive noise and interference , 2003 .

[5]  A. Wilson,et al.  Discharge detection under noisy conditions , 1974 .

[6]  K. Arii,et al.  Separation of noise from partial discharge signals by wavelet , 1997, Proceedings of 5th International Conference on Properties and Applications of Dielectric Materials.

[7]  L. Satish.  Short-time Fourier and wavelet transforms for fault detection in power transformers during impulse tests , 1998 .

[8]  Gerald T. Heydt,et al.  Transient power quality problems analyzed using wavelets , 1997 .

[9]  David L. Donoho,et al.  De-noising by soft-thresholding , 1995, IEEE Trans. Inf. Theory.

[10]  L. Niemeyer,et al.  The importance of statistical characteristics of partial discharge data , 1992 .

[11]  Stéphane Mallat,et al.  A Theory for Multiresolution Signal Decomposition: The Wavelet Representation , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Pawel Zydron,et al.  APPLICATION OF SIGNAL THEORY PROCEDURES TO PARTIAL DISCHARGE DETECTION METHODS , 1995, Proceedings of 1995 IEEE 5th International Conference on Conduction and Breakdown in Solid Dielectrics.

[13]  Xiandong Ma,et al.  Automated wavelet selection and thresholding for PD detection , 2002 .

[14]  E. M. Lalitha,et al.  Wavelet analysis for classification of multi-source PD patterns , 2000 .

[15]  Hong-Tzer Yang,et al.  A de-noising scheme for enhancing wavelet-based power quality monitoring system , 2001 .

[16]  B. Florkowska,et al.  Distortion of partial-discharge images caused by high-voltage harmonics , 2006 .

[17]  E. Gulski,et al.  Computer-aided recognition of discharge sources , 1992 .

[18]  M. L. Hilton,et al.  Data analytic wavelet threshold selection in 2-D signal denoising , 1997, IEEE Trans. Signal Process..