Computer-aided recognition of discharge sources

Gulski, Ε.; Kreuger, F.H.

doi:10.1109/14.123443

Cited by 220 publications

(86 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Several authors [2,3,4] In [5] various descriptors as phase position, magnitude, shape, inception symmetry, pulse distribution, range, density and magnitude consistency are related, while [6,7] use two fractal features (fractal dimension and lacunarity) of the whole image for the pattern recognition.…”

Section: Features Generationmentioning

confidence: 99%

“…2 The limits of this sum must be corrected to include only those coecients which are not subject to to reection conditions and then obtain an unbiased estimator of the variance.…”

Section: Is a Stationary Process (Where D Is A Nonnegativementioning

confidence: 99%

“…Several authors [2,3,4] divide the voltage cycle into phase windows, determine dierent distributions and classify the patterns with statistical operators summarizing these distributions. In [5] the collection is characterized with several features such as phase position, magnitude or shape, while [6,7] use two fractal features of the whole image for the pattern recognition.…”

Section: The Matrix Of Features Which Describes the Collection Of Sigmentioning

confidence: 99%

See 2 more Smart Citations

Signal analysis and feature generation for pattern identification of partial discharges in high-voltage equipment

David¹,

Sánchez-Urán²,

Álvarez³

et al. 2013

Electric Power Systems Research

View full text Add to dashboard Cite

This paper proposes a method for the identication of dierent partial discharges (PD) sources through the analysis of a collection of PD signals acquired with a PD measurement system. This method, robust and sensitive enough to cope with noisy data and external interferences, combines the characterization of each signal from the collection, with a clustering procedure, the CLARA algorithm.Several features are proposed for the characterization of the signals, being the wavelet variances, the frequency estimated with the Prony method, and the energy, the most relevant for the performance of the clustering procedure.The result of the unsupervised classication is a set of clusters each containing those signals which are more similar to each other than to those in other clusters. The analysis of the classication results permits both the identication of dierent PD sources and the discrimination between original PD signals, reections, noise and external interferences. The squared gain function associated with the wavelet lter at scale j.MRA Multiresolution analysis.

show abstract

Section: Features Generationmentioning

confidence: 99%

“…2 The limits of this sum must be corrected to include only those coecients which are not subject to to reection conditions and then obtain an unbiased estimator of the variance.…”

Section: Is a Stationary Process (Where D Is A Nonnegativementioning

confidence: 99%

Section: The Matrix Of Features Which Describes the Collection Of Sigmentioning

confidence: 99%

See 1 more Smart Citation

Signal analysis and feature generation for pattern identification of partial discharges in high-voltage equipment

David¹,

Sánchez-Urán²,

Álvarez³

et al. 2013

Electric Power Systems Research

View full text Add to dashboard Cite

show abstract

“…The first one is formation of so called Feature Vector or Fingerprint and the second one is pattern recognition phase (classification algorithm) itself. Over the last 15-20 years, several PD classification algorithms have been proposed and tested, including statistical tools, signal processing tools, image processing techniques, time-series analysis, fuzzy logic, artificial neural networks (ANN) and hybrid approaches, for both extraction of feature vector and classification [6][7][8][9][10].…”

Section: Introductionmentioning

confidence: 99%

PD Pattern Recognition Using ANFIS

Chalashkanov

Kolev

Dodd

et al. 2008

2008 Annual Report Conference on Electrical Insulation and Dielectric Phenomena

View full text Add to dashboard Cite

This is the accepted version of the paper.This version of the publication may differ from the final published version. Abstract-An application of an adaptive neuro-fuzzy inference system (ANFIS) has been investigated for partial discharge (PD) pattern recognition. The proposed classifier was used to discriminate between PD patterns occurring in internal voids. Three different void shapes were considered in this work, namely flat, square and narrow. Initially, the input feature vector used for classification was based on 15 statistical parameters. The discrimination capabilities of each feature were assessed by applying discriminant analysis. This analysis suggested that some of the features possess much higher discriminatory power than the others. As a result, a simplified classifier with reduced feature vector has been obtained. The results demonstrate the importance in identifying and removing redundancy in the input feature vector for reliable PD identification. Permanent repository link

show abstract

“…Noise rejection techniques, based on the difference of pulse arrival time from two bus couplers [2], [4], pulse-by-pulse noise rejection methods using multiple sensors [3] and denoising method using wavelets [8] have been reported, as the attenuation and frequency content of the propagated PD and noise signals are different. After denoising, the type of developing PDs are identified using the hybrid clustering method, phase-resolved PD patterns, PD magnitude/number distribution, PD trend analysis, spectral analysis, PD-based indexes, and by determining statistical parameters [9], [10]. With single PD, propagation and attenuation characteristics at different electrical lengths [11], [12] are extensively evaluated to identify the location of PD.…”

Section: Introductionmentioning

confidence: 99%

Simulation of Partial Discharge Propagation in Power Network

Birlasekaran

Choi

2005

IEEE Trans. On Energy Conversion

View full text Add to dashboard Cite

Abstract-Locating partial discharge (PD) using single-pulse analysis is considered in this research. The studied network is a generator connected to a transformer through the busbar. Since the stator winding of a generator consists of a finite number of wound coils and finite length busbar, an analysis based on a simple ladder transmission-line model with a finite number of sections has been carried out. A simulation study using the SIMULINK package on the model produces network frequency responses under varied line parameters, and a varied number of sections and termination conditions. The investigation shows that the results of simultaneous PD measurements at different locations can be reproduced satisfactorily using the network model in terms of predicting its resonant frequencies and the time-domain envelope profile of the PD. The proposed method enables one to discriminate PD pulses originating from different locations of the generator winding, as it is observed that the PD magnitude attenuates during propagation.Index Terms-Busbar and transformer, partial-discharge (PD) models of generator, PD propagation, transfer function.

show abstract

Computer-aided recognition of discharge sources

Cited by 220 publications

References 6 publications

Signal analysis and feature generation for pattern identification of partial discharges in high-voltage equipment

Signal analysis and feature generation for pattern identification of partial discharges in high-voltage equipment

PD Pattern Recognition Using ANFIS

Simulation of Partial Discharge Propagation in Power Network

Contact Info

Product

Resources

About