Methods for wavelet-based autonomous discrimination of multiple partial discharge sources

Abstract

Recent years have seen increased interest in the application of on-line condition monitoring of medium voltage networks as the need to maintain and operate ageing cable networks increases. Detection and analysis of partial discharge (PD) activity is generally used as an indicator of pre-breakdown processes that may be indicative of insulation degradation over time. A significant challenge for on-line monitoring is discrimination between multiple partial discharge sources that will often naturally exist in the data. To discriminate between PD sources each PD signal is represented as a feature vector and a clustering algorithm is used to identify clusters in the resulting feature vector space, often after dimensional reduction. Each cluster identified in the data corresponds to a distinct PD source. In this work a comparison of clustering algorithms and dimensional reduction techniques is performed to identify clusters for a variety of PD data sets, in all cases the feature vector is created using wavelet decomposition energies. The three clustering algorithms used were Density Based Spatial Clustering of Applications with Noise (DBSCAN), Ordering Points to Identify Clustering Structure (OPTICS) and Simple Statistics-based Near Neighbour clustering technique (SSNN). The two dimensional reduction techniques considered were Principal Component Analysis (PCA) and t-Distributed Stochastic Neighbour Embedding (t-SNE). At present the most commonly used combination of dimensional reduction technique and clustering algorithm for PD data are PCA and DBSCAN respectively. From the comparison performed in this work it was found that t-SNE combined with OPTICS or SSNN were the most successful at clustering PD data. For use in practical situations SSNN is preferred over OPTICS as it requires only a single input parameter, which generally be hardcoded, leading to a completely autonomous technique. It is therefore suggested that a combination of t-SNE and SSNN is particularly appropriate for discriminating PD sources.

Partial Discharge Electrical Tree Growth Identification by Means of Waveform Source Separation Techniques

“…One of the main advantages these techniques provide is the low computational burden since they are only based on the direct application of simple mathematical expressions. Furthermore, they do not require additional processing before or after the separation, as it happens with other separation techniques which require the prior use of noise filtering techniques on the signals to be classified [16] or the application of dimensionality reduction techniques such as principal component analysis (PCA) [17]- [18], or t-Distributed Stochastic Neighbor Embedding (t-SNE) [19]. These techniques aim to reduce the number of parameters obtained so that the sources representation can be done in a 2D or 3D separation map.…”

Decision tree‐based method for optimum decomposition level determination in wavelet transform for noise reduction of partial discharge signals

“…Partial discharge (PD) happens in the insulation of HV devices because of numerous defects such as cavities, cracks, and electrical trees in the insulation, where their existence is unavoidable [1][2][3][4]. Hence, PD monitoring is one of the most useful assessments of HV equipment insulation before its deterioration and break down [5]. However, the most challenging phenomenon, in PD monitoring, is the noise present in the process of PD signals measuring, which can even completely mask the desired signal [6].…”

Discrimination of three or more partial discharge sources by multi‐step clustering of cumulative energy features

“…Most of the discrimination methods separate mixed signals by clustering the representative features extracted from PD pulse waveforms [5][6][7][8][9][10][11][12][13][14][15][16]. Various kinds of signal analysis algorithms such as wave shape analysis [7,8], fast Fourier transform (FFT) [5,6], spectral-power analysis [9][10][11], wavelet decomposition [12,13], S-transform [14,15] and mathematical morphology (MM)-based signal decomposition [16] were adopted. Among some of these techniques, several kinds of feature selection algorithms such as principal component analysis, t-distributed stochastic neighbour embedding (t-SNE), non-dominated sorting genetic algorithm II and particle swarm optimisation were employed to discover features with optimal separation capability [12][13][14][15].…”

“…Various kinds of signal analysis algorithms such as wave shape analysis [7,8], fast Fourier transform (FFT) [5,6], spectral-power analysis [9][10][11], wavelet decomposition [12,13], S-transform [14,15] and mathematical morphology (MM)-based signal decomposition [16] were adopted. Among some of these techniques, several kinds of feature selection algorithms such as principal component analysis, t-distributed stochastic neighbour embedding (t-SNE), non-dominated sorting genetic algorithm II and particle swarm optimisation were employed to discover features with optimal separation capability [12][13][14][15]. In one of our previous papers, we proposed a self-adaptive separation algorithm by optimised feature extraction of cumulative energy (CE) function, in which the density function (DF) is adopted to evaluate separation capability [17].…”