Search

Today high speed protection of power system elements is necessary and realizable. This paper introduces time-domain protection techniques for transmission lines using voltage and current signals in Cartesian coordinates. Three techniques are proposed using this approach for line protection. Direction of fault is obtained using change in angle between Cartesian planes of voltage and current. In the second application on unit protection scheme, internal and external faults in transmission lines are discriminated using dot product of the normals to both end current planes. Fault type identification is obtained using the angle between incremental changes in currents in Cartesian coordinates. The output of each protection function using the new concept is derived in a quarter-cycle. The performance of each of the proposed techniques is tested for load variation, high resistance fault, different fault types, power swing, fault inception angles and signal with noise. Further, the proposed techniques are evaluated using field data from a 400 kV double circuit line. [ABSTRACT FROM AUTHOR]

Several schemes are available in the literature to protect transmission lines (TLs). These schemes are either transmission network-specific or require modifications in case of any changes in the TL structure. FACTS devices are often incorporated in TLs to increase power transfer and improve controllability and stability. There have been many developments in the field of protection of FACTS-TL. However, the search for a scheme that can successfully adapt to the system changes and can be applied to almost any transmission network to identify the line faults is still ongoing. With advancements in communication technology and global positioning system-enabled time-tagged data, the voltage and current data can be synchronized and, therefore, the pilot relaying may become more viable. Utilizing these technological advances, a digital pilot relaying scheme, for the protection of TLs with or without series compensation and any type of short-circuit fault under any loading condition, is proposed in this paper. [ABSTRACT FROM AUTHOR]

For fault classification in industrial processes, both inaccurate and incomplete supervised information commonly exist in practice, which raise a big challenge to the research field. In this article, a weakly supervised form of the multilayer perceptron (MLP) model is proposed, with considerations of both inaccurate and incomplete labels in fault classification. First, a label probability transition matrix is used to describe the relationship between the inaccurate labels and unknown true labels of process data. This transition matrix is estimated through a Gaussian mixture model, and then used to correct the loss function of the MLP model. Based on the framework of the developed weakly supervised MLP (WS-MLP) model, the information of incomplete labels in the training data set is incorporated simultaneously with the inaccurate label information. The performance of the proposed model WS-MLP is evaluated through two industrial benchmark data sets, results of which indicate its effectiveness under different application cases. Note to Practitioners—Due to manual labeling and other reasons, it is difficult to ensure that the labels of industrial data samples are completely accurate. At present, a useful method is to use a label probability transition matrix to correct the model loss function, so that the model can learn the inaccurate labeled samples robustly. Based on the different distribution of feature representation in the model between accurate category samples and inaccurate category samples, we introduce the Gaussian mixture model to carry out the feature representation and estimate the label probability transition matrix. In addition, a large number of unlabeled samples are available in the industry. In this article, the inaccurate label data samples and incomplete label data samples are modeled using WS-MLP model. The performances of the estimated label probability transition matrix and WS-MLP model are validated by two industrial examples. [ABSTRACT FROM AUTHOR]

Induction motors are important equipment in modern industry. However, the occurrence of fatigue failure following an extended period of operation invariably results in a catastrophic failure. As a result, monitoring and diagnosing induction motors is critical to avoiding unplanned shutdowns caused by premature failures. This article aims to develop an effective method for motor fault detection using time–frequency contents of vibration signals and an attention-based convolutional neural network model. First, the vibration signals are collected and labeled into five different categories: normal condition, outer ring fault, inner ring fault, misalignment condition, and broken rotor bar. Then, using the Morlet function, continuous wavelet transform (CWT) converts the vibratory time-series signals to the scalogram feature images. The time–frequency feature images are created after downsampling and converting the measured vibration signals to the frequency domain. These images are then resized and fed into the proposed convolutional attention neural network (CANN) to identify various induction motor failures. The experimental results demonstrate that the suggested model can provide an excellent diagnosis accuracy of 99.43%, significantly better than the state-of-the-art deep learning approaches for fault diagnosis. Moreover, the developed model’s robustness is validated against adversarial attacks based on the fast gradient sign method (FGSM) by including white Gaussian noise. [ABSTRACT FROM AUTHOR]

Convolutional neural network (CNN) has gained increasing attention in fault classification. However, the performance of CNN is sensitive to its learning rate. Some previous works have been done to tune the learning rate, including the “trial and error” and manual search, which heavily depend on the experts’ experiences and should be conducted repeatedly on every dataset. Because of the variety of the fault data, it is time-consuming and labor intensive to use these traditional tuning methods for fault classification. To overcome this problem, in this article, we develop a novel learning rate scheduler based on the reinforcement learning (RL) for convolutional neural network (RL-CNN) in fault classification, which can schedule the learning rate efficiently and automatically. First, a new RL agent is designed to learn the policies about the learning rate adjustment during the training process. Second, a new structure of RL-CNN is developed to balance the exploration and exploitation of the agent. Third, the bagging ensemble version of RL-CNN (RL-CNN-Ens) is presented. Three bearing datasets are used to test the performance of RL-CNN-Ens. The results show that RL-CNN-Ens outperforms the traditional DLs and machine learning methods. Meanwhile, RL-CNN-Ens can find the state-of-the-art learning rate schedulers as human designed, showing its potential in fault classification. [ABSTRACT FROM AUTHOR]

In this study, analysis of photovoltaic (PV) string currents is performed to understand the behavior of the PV array under faults. The whole analysis is summed up in just two simple statements, and an algorithm is devised on the basis of these statements. The proposed algorithm can detect, classify and localize line-to-ground (L-G) & line-to-line (L-L) faults in the PV array. The proposed method can detect faults of any mismatch level and size of the PV array by using only one current sensor per string. It is also capable of working with finite fault-resistance as well as in the presence/absence of blocking diodes. To check the accuracy of the algorithm, it has been thoroughly tested through experimental setup. [ABSTRACT FROM AUTHOR]

This article proposes the idea of fault classification-based control for the steady-state operation of unmanned surface vehicles (USVs). The idea of fault classification is achieved with the help of wavelet transforms and support vector machines, and the control is performed using a wavelet fuzzy controller. Initially, a brief idea of faults that affect the stable operation of USVs is identified. Furthermore, the surge and sway translational motion of USVs are realized with the help of a ball balancer setup. The fault data are measured in terms of plate angle, ball position, and motor operating voltage for developing the fault classifier. The proposed algorithm depicted improved classification accuracy when compared with conventional methods. To accommodate the operation of the system as per the operating state, a wavelet-based fuzzy controller is proposed. The proposed controller solves the problem of position tracking and balancing for ball and plate system with high precision, hence achieving the stable operation. [ABSTRACT FROM AUTHOR]

The problem of fault classification in industry has been studied extensively. Most classification algorithms are modeled on the premise of data balance. However, the difficulty of collecting industrial data in different modes is quite different. This inevitably leads to data imbalance, which will adversely affect the fault classification performance. This article proposes a novel data augmentation classifier (DAC) for imbalanced fault classification. Data augmentation based on generative adversarial networks (GANs) is an effective way to solve the problem of unbalanced classification. However, the randomness of the GAN generation process restricts the effect of data enhancement. DAC proposes a data selection strategy based on data filtering and data purification in model training to solve this problem. In addition, DAC combines supervised learning and data generation processes to obtain an end-to-end model. Meanwhile, multigenerator structure of DAC (MDAC) is proposed to solve the problem of incomplete learning of a single generator when data imbalances get complicated. The proposed DAC and MDAC are applied in two fault classification cases of the Tennessee Eastman (TE) benchmark process, results of which show superiority of DAC and MDAC compared to existing methods. Note to Practitioners—Data imbalances are common in fault classification and affect the effectiveness of modeling in industry. As a generative model, generative adversarial networks (GANs) provide new ideas for small-class data augmentation. However, the instability of its training process and the randomness of data generation affect the results of data augmentation. In this article, the GAN generation process is analyzed in detail. The results of the visualization indicate that no data generation was perfect at any one time. Based on the rules of GAN data generation, we propose a data selection strategy during training. High-quality data are selected for data augmentation through data filtering and data purification. Apart from this, we combine the training process of GAN and classification model for imbalanced data to reduce modeling time. Through industrial examples, we have evaluated the effectiveness of this method. [ABSTRACT FROM AUTHOR]

In this research work, a diode-based circuit and an algorithm are proposed to detect, classify, and localize faults in small- or medium-scale photovoltaic (PV) array with blocking diodes. Using a single voltage sensor for the whole PV array, the proposed method can detect line-to-line (L-L) and line-to-ground (L-G) faults and find the approximate number of faulty modules. In addition, the proposed circuit uses one voltage sensor to classify faults and one to detect the exact faulty string. Alongside, the proposed algorithm is capable of detecting low and high impedance faults. In contrast to complicated strategies, easy circuit analysis is done to examine the behavior of PV arrays in the event of faults. The proposed algorithm and circuit has been thoroughly tested through a hardware prototype. [ABSTRACT FROM AUTHOR]

Fault classification is vital in smart manufacturing, and convolutional neural network (CNN) has been widely applied in fault classification. But the performance of CNN heavily depends on its learning rate. As the default setting on learning rate cannot guarantee its performance, the learning rate tuning process becomes essential. However, the traditional learning rate tuning methods either cost much time consumption or rely on the experts’ experiences, so it is a considerable barrier for the users. To overcome this drawback, this article proposes a CNN with automatic learning rate scheduler (AutoLR-CNN) for fault classification. First, the long short-term memory (LSTM) is used to extract the features of the past loss of CNN. Then, an agent based on deep deterministic policy gradient (DDPG) is trained to automatically control the learning rate for CNN online. Third, the double CNN structure is developed to enhance the stability of the proposed method. The proposed AutoLR-CNN is tested on two famous bearing data sets and a practical bearing data set on wind turbine. The results of AutoLR-CNN are superior to six commonly used baseline learning rate schedulers in Tensorflow. AutoLR-CNN is also compared with other reported machine learning and deep learning methods. The results show that AutoLR-CNN has achieved the state-of-the-art performance in fault classification. [ABSTRACT FROM AUTHOR]

Deep belief network (DBN) is one of the most representative deep learning models. However, it has a disadvantage that the network structure and parameters are basically determined by experiences. In this article, an improved quantum-inspired differential evolution (MSIQDE), namely MSIQDE algorithm based on making use of the merits of the Mexh wavelet function, standard normal distribution, adaptive quantum state update, and quantum nongate mutation, is proposed to avoid premature convergence and improve the global search ability. Then, the MSIQDE with global optimization ability is used to optimize the parameters of the DBN to construct an optimal DBN model, which is further applied to propose a new fault classification, namely MSIQDE-DBN method. Finally, the vibration data of rolling bearings from the Case Western Reserve University and a real-world engineering application are carried out to verify the performance of the MSIQDE-DBN method. The experimental results show that the MSIQDE takes on better optimization performance, and the MSIQDE-DBN can obtain higher classification accuracy than the other comparison methods. [ABSTRACT FROM AUTHOR]

Fault diagnosis of bearings using classification techniques plays an important role in industrial applications, and, hence, has received increasing attention. Recently, significant efforts have been made to develop various methods for bearing fault classification and the application of Case Western Reserve University (CWRU) data for validation has become a standard reference to test the fault classification algorithms. However, a systematic research for evaluating bearing fault classification performance using the CWRU data is still lacking. This paper aims to provide a comprehensive benchmark analysis of the CWRU data using various entropy and classification methods. The main contribution of this paper is applying entropy-based fault classification methods to establish a benchmark analysis of entire CWRU datasets, aiming to provide a proper assessment of any new classification methods. Recommendations are provided for the selection of the CWRU data to aid in testing new fault classification algorithms, which will enable the researches to develop and evaluate various diagnostic algorithms. In the end, the comparison results and discussion are reported as a useful baseline for future research. [ABSTRACT FROM AUTHOR]

Fault classification of power distribution cables is essential for tripping relays, pinpointing fault location, and repairing failures of a distribution network in the power system. However, existing fault-classification techniques are not totally satisfactory because they may: 1) require the precalibration of responding threshold for each network; 2) fail to identify the three-phase short-circuit faults. since some electrical parameters (e.g., phase angle) are still symmetrical even in abnormal status; and 3) be invulnerable of electromagnetic interferences. In this paper, a fault-classification technique by detecting decaying dc components of currents in faulted phases through magnetic sensing is proposed to overcome the shortcomings mentioned above. First, the three-phase currents are reconstructed by magnetic sensing with a stochastic optimization algorithm, which avoids the waveform distortion in the measurement by current transformers that incurred by the dc bias. Then, the dc component is extracted by mathematical morphology (MM) in phase currents to identify the fault type together with the polarity of dc components. This method was verified successfully for various fault types on a 22-kV power distribution cable in simulation and also a scaled power distribution network experimentally. The proposed method can enhance the reliability of the power distribution network and contribute to smart grid development. [ABSTRACT FROM AUTHOR]

Rotating machinery is widely applied in industrial fields. However, it generally operates under tough working conditions, which leads to the weak fault features and renders fault diagnosis more difficult. In this case, an emerging method called sparse representation classification (SRC) is proposed to enhance the fault features and identify the fault status. However, the typical SRC theory fails to consider the locality of the test sample and training sample, and the training set generally contains much redundant information, which may reduce the fault recognition accuracy. Moreover, the time-shift deviation of vibration signal cannot be avoided effectively using a typical SRC model. To overcome the above-mentioned problems, a novel SRC model, i.e., weighted SRC based on dictionary learning (DL-WSRC), is proposed. For the training set, different fault signals are learned based on the improved K-singular value decomposition (K-SVD) algorithm, which can not only adaptively update the whole training set but also reduce redundant information so as to enhance the sample fault features. For the test sample, DL-WSRC selects an accurate time-domain parameter using the K-means clustering algorithm and computes the weighted coefficients according to the parameter distance between the test sample and the training samples. Then, it sparsely represents the test sample by solving a weighted $l_{0}$ -norm problem. The goal of weighting is to pay more attention to the locality of the sample so as to improve the recognition accuracy. Finally, according to the results of sparse representation, the fault status can be identified through the correlation analysis, which can effectively solve the time-shift deviation problem. The effectiveness of the proposed method is validated by the experiments of rotating machinery, and the results indicate that the proposed method realizes fault classification with a high accuracy. [ABSTRACT FROM AUTHOR]

Online monitoring and diagnostics of permanent magnet synchronous motors (PMSMs) is becoming important with the increasing demand in PMSM applications. Most of the research effort focuses on motor current signature analysis (MCSA) as it can provide remote, online monitoring at low cost. However, all types of defects that produce asymmetries in the PMSM rotor or load produce identical rotor rotational frequency components. This is a serious limitation when applying MCSA since it can produce false indications and degrade the sensitivity of fault detection. The requirement of complex time–frequency analysis techniques is another limitation of MCSA for applications with speed variations. In this paper, the feasibility of using analog Hall sensor signals to complement MCSA for detection and classification of rotor- and load-related defects is investigated. It is shown that Hall sensors installed in machines for initial rotor position estimation can be used with minimal hardware modifications to detect and classify signatures produced by the rotor and load for cases where MCSA fails, even during transient conditions. Experimental testing performed on an interior permanent magnet synchronous motor (IPMSM) under eccentricity, local demagnetization, and load unbalance conditions shows that the reliability and sensitivity of fault detection in PMSM systems can be improved at low added cost. [ABSTRACT FROM AUTHOR]

This paper presents a decision tree based high-speed intelligent scheme for series-compensated transmission line protection using differential phase angle of superimposed current (DPASC). The conventional distance relaying scheme undergoes many shortcomings in transmission line protection in the presence of series compensation devices. The proposed scheme calculates the differential phase angle of current signals using full cycle discrete Fourier transform. The DPASC for each phase is used as an input to the data mining model known as decision tree (DT). The DT is used to generate thresholds to detect and classify the faults since determination of robust thresholds is an important challenge in protective relay engineering. The proposed method has been tested for different operating modes of transmission line in SIMIULINK/MATLAB software. Simulation results indicate that the proposed method is able to detect and classify the faults with acceptable accuracy, in less than half cycle. [ABSTRACT FROM AUTHOR]

Accurate pin-pointing of faults on transmission lines minimizes outage time, labor, and costs. This letter proposes a simple yet effective ensemble Kalman filter (EnKF) approach to accurately locating the faults on transmission lines within half-cycle time. The approach is easy to implement and does not require foreknowledge of either the fault type or an approximate guess of the fault location. Extensive results validate the effectiveness and accuracy of the EnKF-based fault location approach. [ABSTRACT FROM AUTHOR]

The paper presents a method using deep learning framework based on convolution neural network (CNN), for identification and localization of faults of transformer winding under impulse test. The results show that the proposed method outperforms the existing methods significantly. The present scheme eliminates the requirement of separate feature extraction and classification algorithms for the analysis of fault current patterns. A part of the proposed network performs feature learning and the other part classifies the features in a supervised manner. The method is computation intensive but capable of achieving very high degree of accuracy; on an average a margin of more than 7% compared to other published literature till date. [ABSTRACT FROM AUTHOR]

In this paper, a detailed set of reliability prediction methodology for switched reluctance motor drive (SRD) considering system specific control strategy and component fault classification is elaborated from component level to system level. At component level, introduction to SRD and its single switch chopping strategy is presented to capture characteristics of SRD reliability under a certain control strategy. The unique fault modes classification and summary methods which are tailored to SRD are applied for identification of system dominant fault modes. Then at system level, binary models (reliability block diagram and part-count model) and multivariate model (Markov model) are adopted to build systematic SRD reliability model, respectively. Especially in the Markov model, state transition diagram and state probability matrix P(t) are built in detailed description to constitute the graphical and numerical Markov reliability model. Conclusions can be drawn that compared with RBD, Markov model can capture the effect of specific control strategy on system reliability, and further demonstrates the stronger consistency with SRD practical operation. Fault simulation and experiments are conducted to illustrate the influence on system operation state caused by a control strategy. Also, the results verify the state assessment of system performance after component-level fault occurs. [ABSTRACT FROM PUBLISHER]

This paper presents a fault detection and classification scheme for a shunt [static var compensator (SVC)] compensated line. The proposed relaying scheme is based on the concept of superimposed sequence components-based integrated impedance (SSCII). For an internal fault, the magnitude of SSCII is small and for an external fault, it is very large. In the SVC compensated line, a fault forces SVC to vary its impedance. Superimposed components are injected due to this impedance variation, along with the fault components. These SVC-injected components are treated as fault-injected components and, therefore, even the sound phases are detected as faulty phases. To avoid such failure, fault-injected superimposed components have been extracted by using the modified prefault data, which is estimated according to SVC impedance variations. The superimposed components measured using the modified prefault data consist of fault-injected superimposed components only. The proposed scheme has been tested for all types of faults, different values of fault resistances, and several fault locations and SVC locations. The results demonstrate that the proposed scheme successfully detects and classifies the faults. Also, the proposed scheme is robust against variations in fault resistance, source impedance, and SVC location. [ABSTRACT FROM AUTHOR]

Fault detection in solar photovoltaic (PV) arrays is an essential task for increasing reliability and safety in PV systems. Because of PV's nonlinear characteristics, a variety of faults may be difficult to detect by conventional protection devices, leading to safety issues and fire hazards in PV fields. To fill this protection gap, machine learning techniques have been proposed for fault detection based on measurements, such as PV array voltage, current, irradiance, and temperature. However, existing solutions usually use supervised learning models, which are trained by numerous labeled data (known as fault types) and therefore, have drawbacks: 1) the labeled PV data are difficult or expensive to obtain, 2) the trained model is not easy to update, and 3) the model is difficult to visualize. To solve these issues, this paper proposes a graph-based semi-supervised learning model only using a few labeled training data that are normalized for better visualization. The proposed model not only detects the fault, but also further identifies the possible fault type in order to expedite system recovery. Once the model is built, it can learn PV systems autonomously over time as weather changes. Both simulation and experimental results show the effective fault detection and classification of the proposed method. [ABSTRACT FROM AUTHOR]

This paper proposes an approach for a 2-D representation of Shannon wavelets for highly reliable fault diagnosis of multiple induction motor defects. Since the wavelet transform is efficient for analyzing non-stationary and non-deterministic vibration signals, this paper utilizes wavelet coefficients deduced from the Shannon mother wavelet function with varying dilation and translation parameters to create 2-D gray-level images. Using the resulting images and their associated texture characteristics, this paper extracts features by generating global neighborhood structure maps, which are used to extract global image features. The texture features are then used as inputs in one-against-all multi-class support vector machines to identify faults in the induction machine. To evaluate the performance of the proposed approach, it is compared with five conventional state-of-the-art algorithms in terms of classification accuracy. In addition, this paper explores the robustness of the proposed approach in noisy environments by adding white Gaussian noise to the acquired vibration signals. The experimental results indicate that the proposed approach outperforms conventional algorithms in terms of the classification accuracy. Moreover, the proposed approach achieves higher classification accuracy, even in noisy environments. [ABSTRACT FROM PUBLISHER]

This paper presents a traveling-wave-based method for fault classification and localization for three-terminal power transmission systems. In the proposed method, the discrete wavelet transform is utilized to extract transient information from the recorded voltages. Support-vector-machine classifiers are then used to classify the fault type and faulty line/half in the transmission networks. Bewley diagrams are observed for the traveling-wave patterns and the wavelet coefficients of the aerial mode voltage are used to locate the fault. Alternate Transients Program software is used for transients simulations. The performance of the method is tested for different fault inception angles, different fault resistances, nonlinear high impedance faults, and nontypical faults with satisfactory results. [ABSTRACT FROM AUTHOR]

This paper presents a fault classification method with faulted phase selection action for single-circuit transmission lines which is based on the symmetrical components of reactive power. The proposed method does not need any threshold to operate and it is thus a setting-free method which is an exceptional attribute for a protective function. The evaluation of different fault cases reveals the capability of the proposed method. [ABSTRACT FROM AUTHOR]

Explicit fault analysis is the basis of protection for the bipolar HVDC Line. The characteristics of the initial values of traveling waves under various internal fault conditions are investigated on the basis of the symmetrical component analysis. The criteria of fault classification and faulty-pole selection are put forward based on the zero- and positive-sequence backward traveling waves, and an integrated traveling wave-based protection scheme is proposed. The simulations based on real-time digital simulation show that the proposed scheme can detect faults rapidly, determine the fault type effectively, and select the faulty pole correctly. [ABSTRACT FROM AUTHOR]

This paper is the second part of a series of two papers addressing a hybrid framework for achieving fault detection, classification, and location, simultaneously. The proposed framework is formed by a variety of analysis techniques, including symmetrical component analysis, wavelet transforms, principal component analysis, support vector machines, and adaptive structure neural networks. In our previous paper, the mathematical foundation of this framework with numerical results obtained by computer-based simulations has been presented. This paper is devoted to discuss the field-programmable gate-array implementation and experimental results acquired by using real-world scenarios. The hardware implementation of the runtime training technique in the proposed framework is an evolvable hardware tested by the power signals used in a power company transmission network for performance evaluation. The runtime training technique allows the FPGA to have learning and re-training capabilities. The main purpose of this paper is to show the applicability of the proposed framework on a hardware platform and test the framework's robustness and evolvability against noises from the system and measurements. [ABSTRACT FROM AUTHOR]

Re-energizing an overhead distribution feeder safely is a major consideration for a utility's safe work practice. One way to improve the safety is to determine whether the feeder still experiences short circuits before it is energized. In this paper, a novel fault detection technique is proposed to detect if a de-energized distribution system still experiences short-circuit faults. The proposed method involves injecting a thyristor-generated-controllable signal into the de-energized feeder. The feeder voltage and current responses are analyzed to determine if a fault still exists. A thyristor gating control strategy and fault detection algorithm are also developed in this paper to detect all possible types of faults that can occur in a system. The effectiveness of the proposed method has been verified through theoretical analysis, computer simulations, and lab tests. [ABSTRACT FROM AUTHOR]

To ensure safe re-energizing of an overhead distribution feeder after it is de-energized for an extended period, a novel fault detection technique by controlling a thyristor-based device is proposed in a companion paper. The device connected in parallel with a breaker or recloser can inject electrical pulses with adjustable strength for the downstream fault detection in a de-energized system. The proposed method can effectively detect different kinds of asymmetrical faults based on the unbalanced fault currents. However, the unbalanced current-based fault detection scheme is not effective for three-phase symmetrical faults detection. Furthermore, a stalled motor or a shunt-connected capacitor bank in the downstream may also behaves like a short-circuit. Therefore, a fault detection algorithm based on the analysis of the harmonic impedance of the de-energized system is developed in this paper. This method is very effective for the symmetrical fault detection and for distinguishing a stalled motor and capacitor bank from a fault. Extensive lab test results are provided in the paper to verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Most works in power systems event classification concern classifying an event according to the morphology of the corresponding waveform. An important and even more difficult problem is the classification of the event underlying cause. However, the lack of labeled data is more problematic in this second scenario. This paper proposes a framework based on frame-based sequence classification (FBSC), the Alternative Transient Program (ATP), and a public dataset to advance research in this area. As a proof of concept, a thorough evaluation of automatic classification of short circuits in transmission lines is discussed. Simulations with different preprocessing (e.g., wavelets) and learning algorithms (e.g., support vector machines) are presented. The results can be reproduced at other sites and elucidate several tradeoffs when designing the front end and pattern recognition stages of a sequence classifier. For example, when considering the whole event in an offline scenario, the combination of the raw front end and a decision tree is competitive with wavelets and a neural network. [ABSTRACT FROM PUBLISHER]

In this paper, a novel method for fault classification in single-circuit transmission lines is presented. The proposed method needs voltages and currents of only one side of the protected line. After detecting the exact time of fault inception, the fault type is recognized by means of a decision-tree algorithm (DT) which is formerly trained by applying the odd harmonics of the measured signals, up to the 19th. Simulation results have shown that the proposed method can classify the faults in less than a quarter of a cycle with highest possible accuracy. [ABSTRACT FROM PUBLISHER]

In this paper, a novel method for fault classification of double-circuit transmission lines is presented. The proposed method needs voltages and currents of only one side of the protected line. After detecting the exact time of fault inception and calculating the odd harmonics of the measured signals, up to the nineteenth, a decision tree algorithm has been employed for recognition of the intercircuit fault type. Also, the proposed method is extended for classification of crossover faults in these transmission lines. Simulation results have shown that the proposed method can classify the faults in less than a quarter of a cycle with the highest possible accuracy. [ABSTRACT FROM PUBLISHER]

A novel technique for fault detection and classification in the extremely high-voltage transmission line using the fault transients is proposed in this paper. The novel technique, called wavelet singular entropy (WSE), incorporates the advantages of the wavelet transform, singular value decomposition, and Shannon entropy. WSE is capable of being immune to the noise in the fault transient and not being affected by the transient magnitude so it can be used to extract features automatically from fault transients and express the fault features intuitively and quantitatively even in the case of high-noise and low-magnitude fault transients. The WSE-based fault detection is performed in this paper, which proves the availability and superiority of WSE technique in fault detection. A novel algorithm based on WSE is put forward for fault classification and it is verified to be effective and reliable under various fault conditions, such as fault type, fault inception time, fault resistance, and fault location. Therefore, the proposed WSE-based fault detection and classification is feasible and has great potential in practical applications. [ABSTRACT FROM PUBLISHER]

In this paper, a novel virtual-instrument-based monitoring and diagnosis technique that permits the identification of rotor faults in small inverter-fed induction motors is presented and discussed. Since it allows the continuous monitoring of the machine, it is a useful tool for the implementation of advanced maintenance strategies. Through a proper experimental setup, the performance, repeatability, and portability of the proposed algorithm have been evaluated. [ABSTRACT FROM AUTHOR]

This paper presents a fast hardware-efficient logic for fault detection and classification in transmission lines, implemented using a field-programmable gate array (FPGA). The general-purpose SPARTAN3E FPGA was employed for developing the prototype, with all the coding done using a hardware description language (HDL) called very high speed integrated circuit (VHDL). The proposed logic employs only one-terminal current samples and is based on wavelet analysis. Depending on the amount of high frequency components in the current signals after processing, the faults are classified into ten types. The Real Time Windows Target Toolbox of MATLAB was used to apply the current signal inputs to the prototype in real time. An adaptive threshold value is chosen, rather than a fixed threshold in the case of faults involving the ground, to make the classification reliable and accurate. The fault classification time is 6 ms, which is about 1/3 of the power frequency cycle (20 ms). A high level of computational efficiency is achieved as compared to the other wavelet-transform-based algorithms, since only the high frequency details at first level are employed in this algorithm. The validity of the proposed logic was exhaustively tested by simulating various types of faults on a system modeled in the Electromagnetic Transients Program/Alternative Transients Program. The proposed logic was found to be highly reliable and accurate, even in the presence of fault resistance. [ABSTRACT FROM AUTHOR]

Accurate fault classification and swift faulted-phase selection are the bases of protection for transmission systems. This paper presents the algorithm of fault classification and faulted. phase selection based on the initial current traveling wave. The characteristics of various faults are investigated on the basis of the Karenbauer transform. The criteria of fault classification and faulted-phase selection are put forward. In order to extract the traveling wave from postfault signals and to construct the algorithm, wavelet transform technology is adopted. The simulations based on the electromagnetic transients program have been done to verify the performance of the proposed algorithm. [ABSTRACT FROM AUTHOR]

This paper presents an intelligent fault classification approach to power transformer dissolved gas analysis (DGA), dealing with highly versatile or noise-corrupted data. Bootstrap and genetic programming (GP) are implemented to improve the interpretation accuracy for DGA of power transformers. Bootstrap preprocessing is utilized to approximately equalize the sample numbers for different fault classes to improve subsequent fault classification with GP feature extraction. GP is applied to establish classification features for each class based on the collected gas data. The features extracted with GP are then used as the inputs to artificial neural network (ANN), support vector machine (SVM) and K-nearest neighbor ( KNN) classifiers for fault classification. The classification accuracies of the combined GP-ANN, GP-SVM, and GP-KNN classifiers are compared with the ones derived from ANN, SVM, and KNN classifiers, respectively. The test results indicate that the developed preprocessing approach can significantly improve the diagnosis accuracies for power transformer fault classification. [ABSTRACT FROM PUBLISHER]

Two of the most expected objectives of transmission line protection are: 1) differentiating precisely the internal faults from external and 2) indicating exactly the fault type using one end data only. This paper proposes an improved solution based on wavelet transform and self-organized neural network. The measured voltage and current signals are preprocessed first and then decomposed using wavelet multiresolution analysis to obtain the high frequency details and low frequency approximations. The patterns formed based on high frequency signal components are arranged as inputs of neural network #1, whose task is to indicate whether the fault is internal or external. The patterns formed using low frequency approximations are arranged as inputs of neural network #2, whose task is to indicate the exact fault type. The new method uses both low and high frequency information of the fault signal 10 achieve an advanced line protection scheme. The proposed approach is verified using frequency-dependent transmission line model and the test results prove its enhanced performance. A discussion of the application issues for the proposed approach is provided at the end where the generality of the proposed approach and guidance for future study are pointed out. [ABSTRACT FROM AUTHOR]

This paper proposes a novel method for transmission-line fault detection and classification using oscillographic data. The fault detection and its clearing time are determined based on a set of rules obtained from the current waveform analysis in time and wavelet domains. The method is able to single out faults from other power-quality disturbances, such as voltage sags and oscillatory transients, which are common in power systems operation. An artificial neural network classifies the fault from the voltage and current waveforms pattern recognition in the time domain. The method has been used for fault detection and classification from real oscillographic data of a Brazilian utility company with excellent results. [ABSTRACT FROM AUTHOR]

In this paper, we propose a fault diagnosis method for a group of distribution transformers using the graphical information of frequency responses. For a group of transformers with similar capacities and ratings, the patterns of transformer's magnitude plots appear to be almost identical with minor variations of amplitudes and locations of existing peaks and valleys in the plots. Therefore, it is reasonable to establish prototypes composed of several peaks and valleys for magnitude plots and to investigate their changes for the purpose of diagnosis. We select six most significant regions for check regions, and their movements are encoded into a ternary string which is then compared with strings already constructed for various types of faults. Each string-by-string comparison is quantified into correlation values by computing Hamming distances and weighting selected bits. The fault type whose correlation values is largest is determined as a most probable fault in a test transformer. The diagnosis performance of the proposed method is proven to be reliable via experiment data on several types of faults. [ABSTRACT FROM AUTHOR]

The great development in computing power has allowed the implementation of artificial neural networks (ANNs) in the most diverse fields of technology. This paper shows how diverse ANN structures can be applied to the processes of fault classification and fault location in overhead two-terminal transmission lines, with single and double circuit. The existence of a large group of valid ANN structures guarantees the applicability of ANNs in the fault classification and location processes. The selection of the best ANN structures for each process has been carried out by means of a software tool called SARENEUR. [ABSTRACT FROM AUTHOR]

In this paper, a fuzzy-logic-based algorithm to identify the type of faults for digital distance protection system has been developed. The proposed technique is able to accurately identify the phase(s) involved in all ten types of shunt faults that may occur in a transmission line under different fault resistances, inception angle, and loading levels. The proposed method needs only three line-current measurements available at the relay location and can perform the fault classification task in about a half-cycle period. Thus, the proposed technique is well suited for implementation in a digital distance protection scheme. [ABSTRACT FROM PUBLISHER]

One of the major challenges in pattern recognition problems is the feature extraction process which derives new features from existing features, or directly from raw data in order to reduce the cost of computation during the classification process, while improving classifier efficiency. Most current feature extraction techniques transform the original pattern vector into a new vector with increased discrimination capability but lower dimensionality. This is conducted within a predefined feature space, and thus, has limited searching power. Genetic programming (GP) can generate new features from the original dataset without prior knowledge of the probabilistic distribution. In this papa, a GP-based approach is developed for feature extraction from raw vibration data recorded from a rotating machine with six different conditions. The created features are then used as the inputs to a neural classifier for the identification of six bearing conditions. Experimental results demonstrate the ability of GP to discover autimatically the different bearing conditions using features expressed in the form of nonlinear functions. Furthermore, four sets of results--using GP extracted features with artificial neural networks (ANN) and support vector machines (SVM), as well as traditional features with ANN and SVM--have been obtained. This GP-based approach is used for bearing fault classification for the first time and exhibits superior searching power over other techniques. Additionaly, it significantly reduces the time for computation compared with genetic algorithm (GA), therefore, makes a more practical realization of the solution. [ABSTRACT FROM AUTHOR]

Series capacitor protected by metal-oxide varistor and air-gap arrangement imposes problems to line protection and other online decisions. Discrete wavelet transform integrated with a fuzzy logic system is designed for fault classification of a transmission line possessing a series capacitor at the midpoint. The approach uses information obtained from the wavelet decomposition of current signals for faulty phase selection and section identification. Two different FLSs are designed for the two classification objectives in this paper. [ABSTRACT FROM AUTHOR]

The Project On Diverse Software (PODS) was a collaborative software reliability research project whose main objectives were: • To evaluate the merits of using diverse(or n-version) software. • To evaluate the computer-based specification language "X". • To compare the effects of representative high-level and low-level languages on productivity and reliability. In addition, there was a secondary objective to monitor the software development process, with particular reference to the creation and detection of software faults. To achieve these objectives, an experiment was mounted which simulated a normal software development process to produce thee diverse programs to the same requirement. The requirement was for a reactor over-power protection (trip) system. Diversity was ensured by having three independent teams to produce the software, using different specification methods (formal and, informal) and different implementation languages (assembly language and Fortran). This also allowed the comparison of specification methods and programming languages to be made. After careful independent development and testing, the three programs were tested against each other in a special test harness to locate residual faults. All phases of the project were carefully documented for subsequent analysis. The major conclusions for this particular project were that: • Diverse software with majority voting failed less frequently than any individual program, but some common faults did exist at the end of normal software development. • Testing diverse programs "back-to-back" proved to be a powerful method of detecting residual faults. • The residual faults were all related to the specification of requirements, and hence, the requirement specification was the only known cause of common mode failure. [ABSTRACT FROM AUTHOR]