Stator Single-Phase-to-Ground Fault Protection for Bus-connected Powerformers Based on Hierarchical Clustering Algorithm

A new criterion to detect the stator single-phase-to-ground fault for Powerformer is proposed in this paper, which is based on the direction of leakage currents, magnitude of leakage currents, and fault point energy dissipation. New fault characteristics of Powerformer are analyzed and a novel criterion is put forward. The proposed approach detects the ground fault by analyzing the direction, magnitude, and energy of leakage current, which is the difference of zero-sequence current fault component between the neutral and the terminal of Powerformer. The fundamental component and the third-harmonic component of leakage current are combined to realize 100% coverage of fault detection for the stator windings of Powerformer. Simulation results have shown that, under different fault conditions, the new scheme can distinguish internal faults from external faults reliably, and can detect in which machine a stator single-phase-to-ground fault is occurring even if the fault resistance is at 5 kΩ.

Today, with the development of technology, the decision-making capabilities of machines have also increased. With their high analytical skills, computers can easily catch points and relationships that may escape the human eye. Thanks to these capabilities, machines are also widely used in the field of health. For example, many machine learning techniques developed on cancer prediction have been successfully applied. Early detection of cancer is crucial to survival. In the early diagnosis of cancer, the rates of drug treatment, chemotherapy or radiotherapy that the person will be exposed to are significantly reduced and the patient gets through this process with the least amount of wear and tear. Gene Expression Cancer RNA-Seq Dataset was used in this study. This data set includes gene expression values of 5 cancer types (BRCA, KIRC, LUAD, LUSC, UCEC). DNA sequences in the dataset were analyzed using k-means and hierarchical clustering algorithms, which are unsupervised machine learning methods. The aim of the study is to develop a usable machine learning model for early detection of cancer at the gene level. Adjusted Rand Index (ARI), Silhouette Score, and Accuracy metrics were used to evaluate the analysis results. The rand index calculates similarity between clusters by counting the binaries assigned to clusters. The adjusted Rand Index is a randomly adjusted version of the Rand Index. The silhouette score indicates how well a data point fits within its own set among separated datasets. The accuracy metric is obtained as a percentage of correctly clustered data points divided by all predictions. Different connection methods are used in the hierarchical clustering algorithm. These are 'complete', 'ward', 'average' and 'single'. As a result of the study, the accuracy in the k-means algorithm was 0.990, the Adjusted Rand Index was 0.79, and the Silhouette Score was 0.14. Looking at the hierarchical clustering, ward performed the best of the four linkage methods, with an ARI score of 0.76 and a silhouette score of 0.13. As a result of the study, the accuracy of in the hierarchical clustering algorithm was 0.999.

Fault location and distance protection in transmission lines are essential smart grid technologies ensuring reliability of the power system and achieve the continuity of service. The objective of this chapter is to presents an accurate algorithm for estimating fault location in Extra High Voltage (EHV) transmission lines using Artificial Neural Networks (ANNs) for high speed protection. The development of this algorithm is based on disturbed transmission line models. The proposed fault protection (fault detection/classification and location) uses only the three phase currents signals at the one end of the line. The proposed technique uses five ANNs networks and consists of two steps, including fault detection/classification and fault location. For fault detection/classification, one ANN network is used in order to identify the fault type; the fault detection/classification procedure uses the fundamental components of pre-fault and post-fault sequence samples of three phase currents and zero sequence current. For fault location, four ANNs networks are used in order to estimate the exact fault location in transmission line. Magnitudes of pre-fault and post-fault of three phase currents are used. The ANNs are trained with data under a wide variety of fault conditions and used for the fault classification and fault location on the transmission line. The proposed fault detection/classification and location approaches are tested under different fault conditions such as different fault locations, different fault resistances and different fault inception angles via digital simulation using MATLAB software in order to verify the performances of the proposed methods. The ANN-based fault classifier and locator gives high accuracy for all tests under different fault conditions. The simulations results show that the proposed scheme based on ANNs can be used for on-line fault protection in transmission line.

In order to reduce pollution effects, utilities perform costly maintenance jobs that become very excessive and sometimes ineffective because they are not applied on time. To make a more efficient use of human and material resources, it is necessary the use of diagnostic tools. In Mexico, since 1994, the national utility CFE uses monitoring systems for detecting the leakage current in power lines for the diagnosis of outdoor insulation. The leakage current magnitude which causes an insulator flashover due to contamination depends on its shape, material and design. However, insulator flashover occurs around the same leakage current magnitude. Based on laboratory experiments and on field experiences, the Electrical Research Institute (IIE) has classified the leakage current magnitude in five ranges. In the same way, IIE establishes five flashover risk levels that represent some maintenance action that, the user must perform. It is clear that, insulators with different shapes will have different risk levels for the same range of leakage current. Therefore, for the most common insulators used in Mexico, a relationship between the leakage current ranges and flashover risk levels was established. When the leakage current is continuously monitored, it is possible to generate an alarm about the probability of a dangerous flashover risk level; it is also possible to recommend the necessary maintenance actions. This paper summarizes the field and laboratory experiences obtained with the measurement of the leakage currents. Flashover risk levels with maintenance actions associated are also proposed. This classification could be used in the diagnostic of outdoor insulators in power lines.

This paper proposes a novel enhanced transmission line pilot impedance algorithm based on parameter identification, which is defined by the fault component sampling values of voltages and currents at both terminals of this line, and is used to distinguish the internal faults from the external faults. For the external fault, the magnitude of the computing inductance or resistance are greater than the determining data based on the series positive sequence inductance or resistance of the protected line, whereas they are smaller than the aforesaid setting values for the internal faults. The algorithm eliminates the inter-phase coupling caused by the unbalanced grounding fault, so it is the individual phase protection algorithm and possesses the high stability and reliability and immunes to the influence of load state and fault resistance. Performance analysis shows that the algorithm can be used various length lines through the line capacitances time domain compensation and may directly apply the line including the series compensating capacitors. Because the algorithm uses the short data window criterion, it has the short operating time and the little influence of transient components. It builds 1000kV 500km line model on EMTP, uses a laboratory model simulation data of a power system from Lan Zhou Dong to Xian Yang 750kV, executes the simulations under each fault. The simulation results indicate the fact that this algorithm can judge the aforesaid various fault state.

USING VIBRATION MONITORING FOR LOCAL FAULT DETECTION ON GEARS OPERATING UNDER FLUCTUATING LOAD CONDITIONS

Voltage source converters (VSCs) are highly vulnerable to DC fault current. Fast and reliable fault location and protection methods are necessities in VSC-based multiterminal DC (VSC-MTDC) systems for its safe and reliable operation. In this paper, a selective transient-voltage based fault protection method is proposed. The protection method first uses fast overcurrent criterion as the starting element. Then based on the boundary characteristic of the supplemental inductors, the transient-voltage difference between the internal fault and the external fault is used to locate the fault line. And the extraction of the transient-voltage is achieved by wavelet transform. The proposed method is based on local information, it can identify the fault line quickly without communication between protection units at different locations. The methods has no auxiliary directional criterion and additional fault pole selection criterion. Besides, this method can be realized with very low sample rates, and it is rather strong against various fault resistances. Corresponding simulations were carried out, verifying its feasibility and validity.

Hierarchical clustering is of great importance in data analysis. Although there are a number of hierarchical clustering algorithms including agglomerative methods, divisive methods and hybrid methods, most of them are sensitive to noise points, suffer from high computational cost and cannot effectively discover clusters with complex structures. When recognizing patterns from complex structures, humans intuitively tend to discover obvious clusters in dense regions firstly and then deal with objects on the border. Inspired by this idea, we propose a local cores-based hierarchical clustering algorithm called HCLORE. The proposed method first partitions the data set into several clusters by finding local cores, instead of optimizing an objective function through iteration like K-means; then, temporarily removes points with lower local density, so that the boundary between clusters is clearer; after that, merges clusters according to a new defined similarities between clusters; and finally, points with lower local density are assigned to the same clusters as their local cores belong to. The experimental results on synthetic data sets and real data sets show that our algorithm is more effective and efficient than existing methods when processing data sets with complex structures.

The inclined plane tracking and erosion test IEC-60587 is not specified for DC testing. A dc test has been developed from the current ac standard and three formulations of silicone rubber tested. These materials were tested under three voltage levels (2.3, 2.7 and 3.2 kV) for both polarities. Positive dc tests have the highest average and peak leakage current and exhibit a higher degree of surface damage. The observed surface degradation pattern is heavily dependant on polarity. Consistently higher levels of erosion have been observed in the higher voltage positive cases. Erosion of the surface always starts at the bottom electrode, and spreads toward the top electrode. Puncturing of the 6 mm thick samples or deep erosion over more than half the distance between electrodes has only been observed under 2.7 and 3.2 kV positive polarity tests. Results have been analysed using a variety of leakage current analysis techniques going beyond the criteria specified in the original ac standard. The low-frequency behaviour of the leakage current was monitored using a 15 sample per second current recorder. The leakage current magnitude is investigated in a case study and is shown to follow a normal distribution. Reversing the polarity of tests shows the leakage current seen in a test is largely independent of the surface degradation pattern present, but the surface degradation pattern over the first three hours may dictate the morphology of ensuing deep erosion.

In this paper, we present results dealing with the non-uniform pollution carried out under 50 Hz applied voltage on a plane model simulating the 1512 L outdoor insulator largely used by the Algerian Company of Gas and Electric Power (SONELGAZ). Many configurations in non-uniform pollution are studied in the ENP's (Ecole Nationale Polytechnique d'Alger) High Voltage Laboratory in order to analyze the impact of polluted layer distribution on the insulator dielectric performances. The polluted solution has a conductivity of 1.2 mS/cm obtained with distilled water and NaCl. Our investigations are particularly focused on the on line monitoring of both position and width of the contaminated layer. The flashover voltage and the leakage current magnitude have been investigated in order to study the flashover process on this insulating surface. A video apparatus is used to reflect the parallel discharges behavior, appearing when the polluted layer reaches a critical width. Phase angle values between applied voltage and leakage current (LC) signals at the fundamental frequency (50 Hz) are calculated using the Fast Fourier Transform (FFT) spectral analysis. Phase angle measurements indicate that the equivalent impedance of the insulator behaves like RC circuit with a high capacitive effect engendered by the pre-established clean band. This effect decreases when electric discharges occur at a particular voltage level. The Discrete Wavelet Transform (DWT) is adopted for the leakage current decomposition in several time-frequency bands. The STD-MRA (Standard deviation-Multi Resolution Analysis) of these frequency bands is calculated and is employed to choose the most interesting details that detect both position and width increasing of the conducting layer. Reported results show that the pollution surface state and the severity of this conducting layer deposited on insulator surface could be identified from the STD-MRA representation of leakage current frequency bands. It was shown that the high frequency band of the leakage current increases before the final flashover when the polluted layer is located in the middle of the plane model. It was established that a good correlation has been found between the insulator state surface and details of the leakage current obtained through the DWT decomposition. In fact, these details provide relevant information on both position and width of the polluted layer nonuniformly distributed on the insulator surface.

The accuracy of existing hierarchical clustering algorithms in distance measurement is not high, in order to improve it, this paper proposes a condensing hierarchical clustering algorithm based on global distance measurement. This method puts forward the concept of global distance and local distance for the distance measurement, effectively depicting the distance between the data objects, in the merger of the clusters, it also proposes two new concepts: the cohesion within the class and the separability between classes, at the same time, on the basis of these two concepts, it constructs an objective function and maximizes the objective function to reach the purpose that the cohesion within the class be as large as possible, the separability between classes be as small as possible, in order to obtain the optimal clustering results. The experimental accuracy of this algorithm is higher than traditional hierarchical clustering algorithms both on the artificial data and real data sets, which shows that the algorithm has higher clustering validity.

This study investigates the effect of nano- and micro-fillers on water diffusion and leakage current of unidirectional electrically corrosion-resistant (ECR)-grade fiber glass-reinforced epoxy composite insulators. The process of fabrication includes dispersion of nano- and micro-fillers in an epoxy resin using high shear mixing followed by ultrasonication; ECR glass fibers were then embedded into the composite mixture by pultrusion technique. Fillers used were high-purity grade uncoated particles of nano-silica, nano-alumina, micron-sized aluminum trihydrate, calcium carbonate, and magnesium titanate. X-ray diffraction (XRD) and scanning electron microscope (SEM) with energy-dispersive analysis of X-rays (EDAX) were used to investigate crystallinity, morphology, and surface composition of the composites. Dye penetration, water diffusion, and leakage current measurements were performed to evaluate the effect of various fillers on the composites. Results evinced that addition of nano- and micro-fillers caused a reduction in the magnitude of leakage current.

We present a novel pairwise clustering method. Given a proximity matrix of pairwise relations (i.e. pairwise similarity or dissimilarity estimates) between data points, our algorithm extracts the two most prominent clusters in the data set. The algorithm, which is completely nonparametric, iteratively employs a two-step transformation on the proximity matrix. The first step of the transformation represents each point by its relation to all other data points, and the second step re-estimates the pairwise distances using a statistically motivated proximity measure on these representations. Using this transformation, the algorithm iteratively partitions the data points, until it finally converges to two clusters. Although the algorithm is simple and intuitive, it generates a complex dynamics of the proximity matrices. Based on this bipartition procedure we devise a hierarchical clustering algorithm, which employs the basic bipartition algorithm in a straightforward divisive manner. The hierarchical clustering algorithm copes with the model validation problem using a general cross-validation approach, which may be combined with various hierarchical clustering methods. We further present an experimental study of this algorithm. We examine some of the algorithm's properties and performance on some synthetic and ‘standard’ data sets. The experiments demonstrate the robustness of the algorithm and indicate that it generates a good clustering partition even when the data is noisy or corrupted.

Clustering and splitting charging algorithms for large scaled wireless rechargeable sensor networks

A systematic review of fault characteristics and protection schemes in hybrid AC/DC networks: Challenges and future directions