Search

This paper presents a supervised machine learning approach using eight popular classifiers for fault classification in power transmission lines. The classification of faults, indicated by the behavior of the electrical signals associated with them, plays a pivotal role in maintaining the reliability, stability, and security of electrical grids. However, most of the previous studies on fault analysis in power systems relied on proprietary data and lack of standardized benchmarks, hindering the comparison of algorithms and making performance more erratic. Moreover, the nonavailability of labeled data for all types of faults is the most problematic. This paper proposes to perform fault classification on a data set created from a real-time Simulink model to standardize performance and advance research in this area. A new strategy for non-intensive feature extraction is applied using relatively simpler techniques, eliminating computationally expensive techniques such as wavelets. Feature selection through dimensionality reduction techniques is used to improve model performance and more efficient use of computational resources. The performance of the learning algorithms (e.g. Decision Tree, Random Forest, etc.) has been analyzed with various preprocessing techniques (e.g. data scaling, transformation, etc.) and tuning of parameters, focusing on the accuracy and computational time ( $T_{c}$ ), for performance generalization and efficiency. Performing specific operations on data in sequence of steps provided flexibility and adaptability in processing the data, making it easy to train, evaluate, and validate the learning algorithms. The results demonstrated that the proposed scheme can be effectively used for fault classification with high accuracy and significant reductions in $T_{c}$ under various operating conditions. The study also determined the best estimator for each classifier when building and training the classifier models, offering a variety of options. Logistic Regression, Random Forest and Support Vector Machine were the outperforming classifiers and proved their potential for classifying faults in electric power transmission lines.

This study proposes a protection relay using a microcontroller to detect and classify faults in transmission lines based on the Wavelet transform. An experimental model was constructed from an actual 115 kV transmission system prototype. The current signal was observed based on the fault type, phase, and position. Clark’s transform and the discrete Wavelet transform (DWT) were applied to transform signals for analysis. The positive and zero sequences obtained from Clark’s transform were used for fault detection and fault classification, respectively. Moreover, the performances between the DWT and the short-time Fourier Transform (STFT) were compared in terms of accuracy and processing time. In addition, the double-detection technique was used to confirm the accuracy of fault detection. Results show that the proposed method is efficient for fault detection and classification. This finding allows the researcher to choose the appropriate analytical method. Moreover, it can also be used as the basis for overcurrent relay algorithm design in the effort to develop more advanced technologies.

The semiconductor industry, driven by technological advancements, is continuously undergoing process micronization. This micronization has led to an increased complexity in the wafer fabrication process and equipment. Inevitably, this change leads to a rise in defect rates. Furthermore, the traditional manual analysis methods for these defects possess several limitations. Notably, the defect analysis process is time-consuming and heavily relies on the expertise of the workers, making it challenging to achieve consistent analysis results. In modern semiconductor manufacturing sites, a vast amount of sensor data is being generated in real-time from various equipment. These sensor data contain information on the wafer’s condition, the progress of the process, and the operating conditions of the equipment. The purpose of this study is to examine the use of this sensor data in creating a deep learning model that leverages information generated within multi-stage manufacturing processes. The proposed method is designed to automatically classify the defect status of wafers and interpret the causes of these defects. Using the sensor data from two process equipment in semiconductor manufacturing, we achieve better classification performance than other existing methods. Through the interpretation of the classification results, we can identify the sensors and equipment that significantly impact the classification results. The proposed methodology is expected to significantly reduce the effort and reliance on engineers for defect analysis, thus greatly improving production efficiency.

AC microgrids are becoming increasingly important for providing reliable and sustainable power to communities. However, the evolution of distribution systems into microgrids has changed the way they respond to faults and hence their protection requirements. Faults in microgrids could hinder operation stability and damage the system components. The types, locations, and resistances of faults, as well as microgrid operation modes, distributed generation penetration levels, load changes, and system topologies, all affect how the microgrid responds to faults. In order to offer quick restoration and to protect the microgrid components, fault detection and classification are therefore essential for microgrids. In this direction, unconventional methods such as artificial intelligence have been increasing in popularity over the last years. Pattern recognition is a methodology that machine learning as an approach to artificial intelligence is concerned with. The combination of protection with machine learning may be motivating in order to achieve the goal of intelligent operation in the smart grid. In this paper, fault detection, classification and location methods are reviewed for microgrid application. Different methods applied for both fault location and fault classification are being classified by the implemented technique. Such methods are explained and analyzed providing the main advantages and disadvantages of each category. Additionally, the research trends in both fields are analyzed and state–of–the–art methods from each category are thoroughly compared. Finally, the research gaps and future directions are identified.

Based on the signum function (sign of magnitudes) of transient energy, this paper proposes a fault classification approach and algorithm for a monopolar HVDC system. The analytical study here shows that the signum function of variation in transient energy has a zero value individually at both ends on no-fault conditions. Moreover, the summation of the signum functions computed for internal DC faults is zero, whereas the same has a non-zero value for external AC faults. The performance of the proposed algorithm has been extensively evaluated by simulating faults on a CIGRE benchmark system for HVDC monopolar configuration using EMTDC/PSCAD software. Three locations of DC line fault, and AC fault at the two ends of the system have been considered for this evaluation. Five fault resistance values (0, 10, 100, 1000, and 2000 ohms) have been simulated for each fault location. The results conform with the theoretical analysis, and the fault classification by the algorithm is 100% accurate. The time taken to detect and classify a DC fault at the mid-point of the 800-km line is 1.5 ms, and that for line-end faults on the DC line is 3 ms for all values of fault resistance. These results show a marked improvement over those reported earlier in the literature using other techniques. A comparison table is given in the last section to corroborate it.

This paper introduces EngineFaultDB, a novel dataset capturing the intricacies of automotive engine diagnostics. Centered around the widely represented C14NE spark ignition engine, data was collected under controlled laboratory conditions, simulating various operational states, including normal and specific fault scenarios. Utilizing tools such as an NGA 6000 gas analyzer and a USB 6008 data acquisition card from National Instruments, we were able to monitor and capture a comprehensive range of engine parameters, from throttle position and fuel consumption to exhaust gas emissions. Our dataset, comprising 55,999 meticulously curated entries across 14 distinct variables, provides a holistic picture of engine behavior, making it an invaluable resource for automotive researchers and practitioners. For evaluation, several classifiers, including logistic regression, decision trees, random forests, support vector machines, k-nearest neighbors, and a feed-forward neural network, were trained on this dataset. Their performance, under standard configurations and a simple neural network architecture, offers foundational benchmarks for future explorations. Results underscore the dataset’s potential in fostering advanced diagnostic algorithms. As a testament to our commitment to open research, EngineFaultDB is freely available for academic use. Future work involves expanding the dataset’s diversity, exploring deeper neural architectures, and integrating real-world automotive conditions.

In industries, squirrel cage induction motors are crucial for supplying rotary motion in power tools. This research presents a robust but simple framework for an inter-turn fault classification at minor loading across diverse fault occurrence conditions, which is one of the most common defects in a squirrel cage induction motor. Early detection of this issue is critical to prevent the system from completely failing as a result of it evolving to a more severe stator winding fault. This study employs a hybrid feature selection strategy (a hybrid of a filter-based and a wrapper-based approach) using the Hilbert Transform signal processing technique and a statistical feature extraction approach, which is then fed to a support vector machine as the classifier. The suggested framework is tested and validated against other known classifier models. The results demonstrate that the model has a computationally low diagnostic performance process with exceptional accuracy. Furthermore, when compared to there classifier models, the suggested model provided the best diagnostic outcome on the stator winding fault classification, demonstrating its dependability in fault diagnostic classification for squirrel cage induction motors.

Traditional fault classification methods typically rely on manual feature extraction and the application of machine-learning algorithms. However, these approaches encounter difficulties when extracting features and handling large-scale datasets. This study proposes a data preprocessing method for accurately detecting various types of short-circuit faults in power systems, which can lead to more effective power repair and maintenance processes. The proposed method involves converting the measured voltage and current signals into time and frequency domains using the short-time Fourier transform (STFT) to produce a time-frequency energy map. A convolutional neural network (CNN) is subsequently trained and tested to classify the short-circuit faults. However, overfitting may occur during the CNN training process owing to the large volume of data with similar features. To address this issue, this study proposes a data reduction method based on the fast dynamic time warping (Fast-DTW) algorithm, which compares waveform features and eliminates highly similar data regarded as redundant data from the dataset. The simulation results show that the proposed method can improve the model training performance and its adaptability to different power system topologies, as tested in two simulation environments: power systems computer-aided design (PSCAD)/electromagnetic transients, including DC (EMTDC), and the real-time digital simulator (RTDS). The STFT transformation is implemented in MATLAB. The simulation results demonstrate that the proposed method reduces redundant data by 40.2%, while decreasing the model training time. Consequently, the overall accuracy, precision, recall and F1 score of the fault classification reaches 99.37%, 99.36%, 99.35% and 99.35%, confirming the effectiveness of the proposed method for fault classification.

In this article, a maiden attempt have been taken for the online detection of faults, classification of faults, and identification of the fault locations of a grid-connected Micro-grid (MG) system. A deep learning algorithm-based Long Short Term Memory (LSTM) network is proposed, for the first time, for the online detection of faults and their classifications of the considered MG system to overcome the issues that persist in the existing algorithms. Also, a combination of an LSTM network and feed-forward neural network (FFNN) with a back-propagation algorithm (BPA) is proposed to identify the exact locations of the faults since the identification of fault locations is more challenging than fault categorizations. To select a suitable deep learning network with multiple hidden layers for achieving the aforesaid objectives, a rigorous analysis has been done. To study the accuracy of the proposed techniques, different types of faults with different parameters are considered in this paper. An extensive simulation has been done in MATLAB/Simulink platform to study the performance of the system with the proposed techniques. To validate the effectiveness of the proposed techniques, the entire system is implemented in the real-time platform using the OPAL-RT digital simulator. Comparison has also been done for the results obtained using ANN and proposed techniques. The results show that the proposed techniques based on the deep learning network effectively detect, classify, and identify the location of different faults of an MG system with acceptable performances.

The technology advancement is supported by additive manufacturing industries, especially 3D printing companies, that enable fast object prototyping and development in Industry 4.0. As 3D printed products are highly adopted in various fields, the final printed product must fulfill precise requirements without any defects. Therefore, an efficient framework that simultaneously learns and detects faults during the printing process is required. Unfortunately, most state-of-the-art studies utilize a centralized approach, which is inefficient for continuous model updates. This article presents a hierarchical federated transfer learning (HFTL) framework that employs edge, fog, and cloud concepts to enable an efficient model updating mechanism. The proposed HFTL uses a well-known DL model with a new classifier to significantly reduce the distributed training process while providing high detection and classification performance. Additionally, the fog server comprehensively exploits the data collection from several edge servers and performs local training. The extensive simulation results indicate that the performance of the proposed HFTL is more efficient, with 24% faster training time, effectively detects flaws in 3D printing products with 45% accuracy, and has a 59% F1-score improvement in non-IID data distribution compared to traditional FL architecture.

Today, the classification and location of faults in electrical networks remains a topic of great interest. Faults are a major issue, mainly due to the time spent to detect, locate, and repair the cause of the fault. To reduce time and associated costs, automatic fault classification and location is gaining great interest. State-of-the-art techniques to classify and locate faults are mainly based on line-impedance measurements or the detection of the traveling wave produced by the event caused by the fault itself. In contrast, this paper describes the methodology for creating a database and a model for a complex distribution network. Both objectives are covered under the paradigm of the time-domain pulse reflectometry (TDR) principle. By using this technique, large distances can be monitored on a line with a single device. Thus, in this way a database is shared and created from the results of simulations of a real and complex distribution network modeled in the PSCADTM software, which have been validated with measurements from an experimental test setup. Experimental validations have shown that the combination of the TDR technique with the modeling of a real network (including the real injector and the network coupling filter from the prototype) provides high-quality signals that are very similar and reliable to the real ones. In this sense, it is intended firstly that this model and its corresponding data will serve as a basis for further processing by any of the existing state-of-the-art techniques. And secondly, to become a valid alternative to the already well-known Test Feeders but adapted to work groups not used to the electrical world but to the environment of pure data processing.

In this paper, special attention is paid to the detection and diagnosis of various incipient faults of uncertain wind energy conversion (WEC) systems. The proposals will enhance the monitoring and diagnosis of the WEC system while taking into account the system uncertainties. The developed techniques are based on Support Vector Machine (SVM) model to improve the diagnosis of WEC systems. First, to deal with model uncertainties in WEC systems, the SVM will be extended to interval-valued data with the aim of achieving greater accuracy and robustness for these uncertainties. Then, to improve even more the performances of the developed interval-valued SVM, multiscale data representation will be used to develop multiscale extensions of interval-valued SVM. Next, as a feature selection tool, an improved extension of Artificial Butterfly Optimization (ABO) algorithm is used in order to extract the significant features from data and improve the diagnosis results of multiscale interval SVM. The proposed improved ABO method consists in reducing the number of samples in the training data set using the Euclidean distance and extracting the most significant features from the reduced data using ABO algorithm. This in turn plays a vital role in improving the accuracy using the multiscale interval-SVM method and reducing the computational and storage costs.

Fault detection and classification are fundamental requirements of multilevel inverters (MLIs) to ensure constant operation and improved reliability. Nowadays, the machine learning (ML) technique is utilized for fault diagnosis in MLIs due to its inherent features such as high accuracy, reduced computation time, and complexity. However, the rich availability of parts and classifiers in ML techniques demands a tedious investigation of every combination to design an optimal fault classifier. To overcome this problem, a combined optimiser is proposed to automate the ML-based classifier design, which involves selecting optimal features and classifier. Mean, total harmonic distortion, root mean square, and different harmonic orders (upto 12th order) of the output voltage of MLI are considered as features and four different ML techniques like K-nearest neighborhood, decision tree, Naive Bayesian classifier, and support vector machines are considered. Ant Colony Optimization (ACO) is used to formulate a combinatorial optimization routine to maximize classification accuracy by optimal selection of features and classifier. The proposed technique is used to design a fault classifier for two different MLIs, such as Cascaded H-bridge MLI (CHBMLI) and Packed U cell inverter (PUC), during the fault conditions (open circuit and short circuit) to check the feasibility. Simulation results illustrate classification accuracy of 97.84% and 98.61% for CHBMLI and PUC, respectively. Experimental validation of the designed classifier on the inverter prototype is also carried out and illustrates 95.56% and 94.28% classification accuracy for CHBMLI and PUC, respectively.

Fault diagnosis and classification (FDC) is an important part of prognostics and health management for ensuring safety and performance in the flight. However, it is challenging to achieve accurate FDC only based on single senor readings. In this paper, a fused FDC model among multiple different sensors is stabled by a hybrid deep learning architecture combining a sparse autoencoder (SAE) and a convolutional neural network (CNN). The hybrid model uses the SAE to enhance the hidden fault signal features in the multiple sensor signals, and then classifies the obtained feature map using the CNN. This method, which combines the advantages of the SAE in feature extraction and of the CNN in local feature recognition, fully utilizes the spatiotemporal coupling characteristics of multi-sensor signals. The FDC accuracy obtained by the proposed method when applied to a flight test data set is 93.78%, compared with 66.67% obtained using the combined SAE and feedforward neural network method and 83.11% obtained using the CNN only.

This study proposes a new method for detecting and classifying faults in distribution lines. The physical principle of classification is based on time-domain pulse reflectometry (TDR). These high-frequency pulses are injected into the line, propagate through all of its bifurcations, and are reflected back to the injection point. According to the impedances encountered along the way, these signals carry information regarding the state of the line. In the present work, an initial signal database was obtained using the TDR technique, simulating a real distribution line using (PSCAD™). By transforming these signals into images and reducing their dimensionality, these signals are processed using convolutional neural networks (CNN). In particular, in this study, contrastive learning in Siamese networks was used for the classification of different types of faults (ToF). In addition, to avoid the problem of overfitting owing to the scarcity of examples, generative adversarial neural networks (GAN) have been used to synthesise new examples, enlarging the initial database. The combination of Siamese neural networks and GAN allows the classification of this type of signal using only synthesised examples to train and validate and only the original examples to test the network. This solves the problem of the lack of original examples in this type of signal of natural phenomena which are difficult to obtain and simulate.

In this study, an integrated framework involving state classification, preprocessing, and classification is proposed for the fault diagnosis of elevator doors using control state information. During state classification, the door state is classified as operational or non-operational; moreover, based on the state information of the control board of an elevator door, the corresponding operational conditions are classified as opening or closing states. In the preprocessing phase, data processing and interpolation and feature manipulation are performed. Data are interpolated to synchronize each measurement with respect to the reference time; then, the state information is manipulated to create distinct features. In the classification phase, the operational states are classified, and nonlinear coordinate transformation is executed to transfer several features into a new nonlinear hyperspace by using an autoencoder. Two manifold features are extracted from the latent layer of the autoencoder; these become the principal axes for state classification using a support vector machine, indicating that three states are possible: normal with and without completion of a full stroke and abnormal stroke. The effectiveness of the proposed method was verified by performing field measurements on a control board during the operation of an elevator. Specifically, the autoencoder in the integrated framework effectively transformed an original space into a highly nonlinear yet efficient manifold, where classification was easy for different operational states. The proposed framework can be effective for real-world detection of elevator door faults because it exclusively uses state information measured from control boards.

This paper investigates the challenging fault prediction problem in process industries that adopt autonomous and intelligent cyber-physical systems (CPS), which is in line with the emerging developments of industrial internet of things (IIoT) and Industry 4.0. Particularly, we developed an end-to-end deep learning approach based on a large volume of real-time sensory data collected from a chemical plant equipped with wireless sensors. Firstly, a novel recursive architecture with multi-lookback inputs is proposed to perform autoregression on imbalanced time-series data as a preliminary prediction. In this process, a novel learning algorithm named recursive gradient descent (RGD) is developed for the proposed architecture to reduce cumulative prediction uncertainties. Subsequently, a classification model based on temporal convolutions over multiple channels with decay effect is proposed to perform multi-class classification for fault root cause identification and localization. The overall network is named the cumulative uncertainty reduction network (CURNet), for its superior capacity in reducing prediction uncertainties accumulated over multiple prediction steps. Performance evaluations show that CURNet is able to achieve superior performance especially in terms of fault prediction recall and fault type classification accuracy, compared to the existing techniques.

Nuclear power plants (NPPs) are complex dynamic systems with multiple sensors and actuators. The presence of faults in the actuators and sensors can deteriorate the system’s performance and cause serious safety issues. This calls for the development of fault detection and diagnosis systems for detection and isolation of such faults. In this study, fault detection and diagnosis (FDD) based on neural networks (NN) and K-nearest neighbour (KNN) algorithm is applied to a pressurized water reactor (PWR). Fault detection is first determined based on the NN. Second, the KNN algorithm is used to classify the faults. The proposed approach is capable of classifying a variety of actuator faults, sensor faults, and multiple simultaneous actuator and sensor faults. A set of simulation results is provided to demonstrate the accuracy of the FDD method. The classifier performance is further compared with other machine learning techniques.

Microgrids (MGs) suffer from unpredictable faults on the feeders for various random reasons. These faults could obstruct the stability of the MG operation and damage the components. Moreover, many uncertainties elements affect the MG’s response to faults, such as faults’ types and locations and resistances, MG operation modes, DG penetration levels, load variations, and system topologies. Therefore, fault detection, classification, and location are vital for the MGs as they provide rapid restoration and protect the components. This paper proposes an adaptive protection (AP) scheme for the future renewable electric energy delivery and management (FREEDM) system. The proposed scheme is based on the convolution neural network (CNN), in which the measured current and voltage at buses are processed in multidimensional arrays for the images’ identification and classification. The gorilla troops optimization (GTO) technique has been used to improve the CNN by acquiring the optimal architecture and hyperparameters of the proposed CNN. The proposed protection scheme can detect the system fault, classify the fault type, and determine the fault location using three proposed CNN-GTO protection scheme models. A communication channel has been performed to transfer the data, information, and tripping signals between the different devices in the FREEDM system. The proposed method is tested using a hypothetical FREEDM microgrid system under different fault conditions. The results show that the proposed CNN-GTO models can detect, classify, and location feeder faults in the FREEDM system with high accuracy. A comparison with the existing schemes such as Support vector machine, Fuzzy logic, conventional CNN, and wavelet-based CNN is performed. The optimized CNN-based GTO models can achieve an overall accuracy for fault detection, classification, and location of 99.37, 99, and 98.2%, respectively.

Fault analysis (detection, classification and location) of transmission network is of great importance in power system. A Wavelet-Alienation-Neural (WAN) technique has been developed for the fault analysis of Unified Power Flow Controller (UPFC) compensated transmission network. The detection and classification of various outages are accomplished by alienation of wavelet based approximate coefficients computed from current signals. The precise location of faults is carried out by an Artificial Neural Network fed from estimated approximate coefficients computed from voltage and current signals of the same quarter cycle. The robustness of the algorithm is proved with the case studies of varying fault locations, sampling frequency, system parameters, effects of noise, fault incipient angle, different control strategies and fault path impedances.

This paper proposes a novel fault detection and diagnosis (FDD) technique for grid-tied PV systems. The proposed approach deals with system uncertainties (current/voltage variability, noise, measurement errors,…) by using an interval-valued data representation, and with large-scale systems by using a dataset size-reduction framework. The failures encompassed in this study are the open-circuit/short-circuit, islanding, output current sensor, and partial shading faults. In the proposed FDD approach, named interval reduced kernel PCA (IRKPCA)-based Random Forest (IRKPCA-RF), the feature extraction and selection phase is performed using the IRKPCA models while the fault classification is ensured using the RF algorithm. The main contribution of the proposed approach is to provide a good trade-off between low computation time and high classification metrics. The performance of the proposed IRKPCA-RF approach is assessed using a set of emulated data of a grid-tied PV system operating under healthy and faulty conditions. The presented results show that the proposed IRKPCA-RF approach is characterized by enhanced diagnosis metrics, classification rate, and computation time compared to the classical techniques.

Accurate and fast fault monitoring system is important for photovoltaic (PV) systems to reduce the damage and energy loss associated with faults. This paper presents a fast fault detection and classification scheme for PV systems. The rate of change of the measured power at the array level is firstly used to detect and differentiate between shading and PV system short-circuit faults. Then, the measured cross-strings differential currents classify the fault type in the case of short-circuit faults. The proposed method accounts for intra-string, string-to-negative terminal, cross-string (string to string), and pole-to-pole short-circuit faults, as well as open circuit faults. Two current transducers per two strings are only required to apply the method leading to a significant reduction in the number of transducers/sensors compared to the literature, making it a cost-effective solution. The fault is detected and classified in 1 ms. A 400 kW PV system consisting of 4 arrays simulated using MATLAB/Simulink package is used for the evaluation process. Short-circuit faults are imposed at different conditions of mismatch levels and fault resistance values. Also, operation at reduced irradiance level is considered, and shading faults are simulated in different levels. The simulation results prove the ability of the proposed scheme to detect and differentiate between shading and short-circuit faults. The proposed method can correctly classify different fault types in almost all cases. The low implementation cost due to the lower number of required sensors and the fast-operating time (1 ms) boost the feasibility of the proposed method.

In this paper, a novel relaying scheme is proposed for bipolar line commutated converter (LCC) high voltage direct current (HVDC) transmission lines that detects the fault, identifies the pole of fault and estimates the location of the fault. The scheme uses features extracted from rectifier end DC current and voltage signals. Long short term memory (LSTM), a deep learning method has been designed as classifier as well as predictor for carrying out different relaying tasks. Three modules have been designed namely LSTM-FD (LSTM module for fault detection (FD)), LSTM-FI (LSTM module for fault pole identification (FI)) and LSTM-FL (LSTM module for fault location estimation (FL)). The voltage and current signals are obtained from measuring units. Then with a moving window of one cycle, the RMS of signals is calculated. The current and voltage features are obtained in the time domain. The voltage and current features obtained are used as input to fault detection and fault pole identification module. For fault location estimation, half cycle samples of RMS current and voltage after the fault have been taken. From half cycle sample, maximum value of current and minimum value of voltage has been obtained. This single value of current and voltage are used as input feature to the fault location module. All the relaying modules have been tested varying fault types, locations, resistances, smoothing reactors, noisy signals, etc. Sensitivity and reliability of the proposed relaying scheme is 100% with all the tested fault cases. Error in location estimation is within 1% for all the tested fault cases. Advantage of the method is that it does not require communication link. Another advantage of the proposed method is that it can work with low sampling frequency. Additionally, reliability and sensitivity of proposed method is very high, hence can be used as an alternative to travelling wave based relaying schemes.

The main objective of this article is to develop an enhanced ensemble learning (EL) based intelligent fault detection and diagnosis (FDD) paradigms that aim to ensure the high-performance operation of Grid-Connected Photovoltaic (PV) systems. The developed EL based techniques consist in combining multiple learning models instead of using a single learning model. To do that, three EL-based FDD techniques are proposed. First, an EL technique that merges the benefits of Support Vector Machine (SVM), K-Nearest Neighbour (KNN), and Decision Tree (DT) is presented. The developed method contributes to the reduction of the overall diagnosis error and has the ability to combine various models. However, classical EL models ignore the time-dependence of PV measurements. In addition, the PV system data are frequently time-correlated. Therefore, kernel PCA (KPCA)-based EL and reduced KPCA (RKPCA)-based EL techniques are developed to take into consideration the dynamic and multivariate natures of the PV measurements. The two proposed KPCA -based EL and RKPCA-based EL techniques are addressed so that the features extraction and selection phases are performed using the KPCA and RKPCA models and the sensitive and significant characteristics are transmitted to the EL model for classification purposes. The presented results prove that the proposed EL based methods offer enhanced diagnosis performances when applied to PV systems.

Insulator failure is one of the important causes of railway power transmission accidents. In the automatic detection system of railway insulators, the detection and classification of insulator faults is a challenging task due to the complex background, small insulators and unobvious failures. In this article, we propose a railway insulator fault detection network based on convolutional neural network, which can detect faulty insulators from images with high resolution and complex background. The insulator fault detection network realizes the position detection and fault classification of the insulator by cascading the detection network and the fault classification network. The method of cascading two networks can reduce the amount of network calculations and improve the accuracy of fault classification. The insulator detection network uses low-resolution images for position detection, and this method can prevent the detection network from paying too much attention to the details of the image, thereby reducing the amount of network calculations. The fault classification network uses high-resolution insulator images for fault classification. The high-resolution images in this method have rich detailed information, which helps to improve the accuracy of fault classification. The trained insulator detection network and the fault classification network are cascaded to form an insulator fault detection network. The precision, recall and mAP values of the insulator fault detection network are 94.10%, 92.88% and 93.46% respectively. Experiment shows show that this network cascading method can significantly improve the accuracy and robustness of insulator fault detection.

This paper proposes a modified relevance vector machine with slow feature analysis fault classification for industrial processes. Traditional support vector machine classification does not work well when there are insufficient training samples. A relevance vector machine, which is a Bayesian learning-based probabilistic sparse model, is developed to determine the probabilistic prediction and sparse solutions for the fault category. This approach has the benefits of good generalization ability and robustness to small training samples. To maximize the dynamic separability between classes and reduce the computational complexity, slow feature analysis is used to extract the inner dynamic features and reduce the dimension. Experiments comparing the proposed method, relevance vector machine and support vector machine classification are performed using the Tennessee Eastman process. For all faults, relevance vector machine has a classification rate of 39%, while the proposed algorithm has an overall classification rate of 76.1%. This shows the efficiency and advantages of the proposed method.

Transmission line fault detection is a complex, high-cost and time-consuming work for deep-sea offshore wind farms. Therefore, rapid intelligent on-line fault detection and classification of submarine transmission lines play a very important role in the offshore wind farm reliability improving and operating costs reduction. This paper presents a novel hybrid on-line detection method that combines Wavelet noise Reduction, Clarke transform, Stockwell transform and Decision Tree (WRC-SDT). First, the measured Wind Turbine (WT) voltage signals are processed through wavelet noise reduction and Clarke transform to get the gradient of the voltage component. The gradient of the voltage component is then monitored in order to detect faults. Second, the recorded WT current date are processed through Stockwell transform with a view to obtaining the transmission line fault eigenvalues. The fault eigenvalues are then taken as the input of decision tree in order to classify different types of faults. To verify the feasibility and performance of the proposed method, the comparison of a detection and classification result is presented based on more than 1600 fault simulation data. The results show that WRC-SDT method is immune to fault resistance, starting angle and location. The proposed method is also robust to measurement noise.

Fault diagnosis in photovoltaic (PV) arrays is essential in enhancing power output as well as the useful life span of a PV system. Severe faults such as Partial Shading (PS) and high impedance faults, low location mismatch, and the presence of Maximum Power Point Tracking (MPPT) make fault detection challenging in harsh environmental conditions. In this regard, there have been several attempts made by various researchers to identify PV array faults. However, most of the previous work has focused on fault detection and classification in only a few faulty scenarios. This paper presents a novel approach that utilizes deep two-dimensional (2-D) Convolutional Neural Networks (CNN) to extract features from 2-D scalograms generated from PV system data in order to effectively detect and classify PV system faults. An in-depth quantitative evaluation of the proposed approach is presented and compared with previous classification methods for PV array faults - both classical machine learning based and deep learning based. Unlike contemporary work, five different faulty cases (including faults in PS - on which no work has been done before in the machine learning domain) have been considered in our study, along with the incorporation of MPPT. We generate a consistent dataset over which to compare ours and previous approaches, to make for the first (to the best of our knowledge) comprehensive and meaningful comparative evaluation of fault diagnosis. It is observed that the proposed method involving fine-tuned pre-trained CNN outperforms existing techniques, achieving a high fault detection accuracy of 73.53%. Our study also highlights the importance of representative and discriminative features to classify faults (as opposed to the use of raw data), especially in the noisy scenario, where our method achieves the best performance of 70.45%. We believe that our work will serve to guide future research in PV system fault diagnosis.

The permanency of highly-reliable power supply is a core trait of an electric power transmission network. A transmission line is the main part of this network through which power is transmitted to the utility. These lines are often damaged by accidental breakdowns owing to different random origins. Hence, researchers are trying to detect and identify these failures at the earliest to avoid financial losses. This paper offers a new real-time fast mathematical morphology-based fault feature extraction scheme for detection and classification of transmission line faults. The morphological median filter is exploited to wrest unique fault features which are then fed as an input to a decision tree classifier to classify the fault type. The acquired graphical and numerical results of the extracted features affirm the potency of the offered scheme. The proposed scheme is verified for different fault cases simulated on high-voltage transmission line modelled using ATP/EMTP with varying system constraints. The performance of the stated technique is also validated for fault detection and classification on real-field transmission lines. The results state that the proposed method is capable of detecting and classifying the faults with adequate precision and reduced computational intricacy, in less than a quarter of a cycle.

Raw statistical features can imitate the amplitude, average, energy and time, and frequency series distribution of a raw vibration signal. However, these raw statistical features are either not very sensitive to weak incipient faults or are unsuitable for more severe faults, thus affecting the fault detection and classification accuracy. To tackle this problem, this paper proposes a discriminant feature extraction method for Centrifugal Pump (CP) fault diagnosis. In order to obtain the discriminant feature pool, the proposed method is divided into three phases. In the first phase, a healthy baseline signal is selected. In the second phase, the healthy baseline signal is cross-correlated with the CP vibration signals of different classes, and a set of new features are extracted from the resulting correlation sequence. In the third phase, raw hybrid features in time, frequency, and the time-frequency domain are extracted from both the healthy baseline signals and the CP vibration signals of different classes. The correlation coefficient is calculated between the raw hybrid feature pools, which results in a new set of discriminant features. Discriminant features help the machine learning classifiers to effectively detect and classify the data into its respective classes. Furthermore, the proposed method combines all these features into a single feature vector that forms a vulnerable feature pool. The vulnerable feature pool describes the CP's vulnerability to a fault and is provided as an input to a multiclass support vector machine (MSVM) for CP fault detection and classification. The experimental results illustrate that the accuracy obtained from the proposed method shows promising improvements over the state-of-the-art conventional methods.

It is always a serious challenge to deal with the influence of noise for fault detection of rotating electrical machines under varying operating environments. To extract the most discriminative features for machine fault classification, a novel dimensionality reduction algorithm called joint sparsity and collaboration preserving projections (JSCPP) is proposed in this article, based on the platform of graph embedding framework. A joint combination of sparse representation (SR) and collaborative representation (CR) is used to construct the intrinsic and penalty graphs. SR emphasizes only on the dictionary atoms highly correlated to the query sample for local optimum while CR takes every atom into consideration for global optimum, thus applying a joint combination would make it benefit from the merits of both for graph construction. After JSCPP seeks the optimal projection directions by minimizing the intra-class compactness and maximizing the inter-class separability, a nearest neighbor classifier is then used. Through two sets of vibration data with white Gaussian noise interfered, the capability of JSCPP is demonstrated by effectively classifying different defect types and severity degrees of faulty bearings. The experimental results have also proved that the classification performance of the proposed method is much superior to the other compared algorithms in both accuracy rates and reliability.

Bearings are vital automation machine elements that are used quite frequently for power transmission and shaft bearing in rotating machines. The healthy operation of the bearings directly affects the performance of the rotating machines. Bearing faults may cause more vibration than normal in rotating machines, which wastes power. However, further bearing failures can cause vital damage to rotating machines. In this study, bearing vibration values are obtained through a special test setup. Different types and different sizes of artificial faults have been created in the bearings for the testing process. Data on these bearings are collected at different speeds. The purpose of the study is to diagnose faults in the bearings. In this context, a new approach is proposed. First, the one-dimensional local binary pattern (1D-LBP) method is applied to vibration signals, and all signal data are carried to the 1D-LBP plane. Statistical features are obtained from the signals in the 1D-LBP plane by using these features, and then the vibrational signals are classified by the gray relational analysis (GRA) model. Four different data sets are organized to test the proposed approach. The results of the test process with this proposed model have an accuracy of 99.044% for Dataset1 (different speed -300 rpm intervals), 94.224% for Dataset2 (different speed -60 rpm intervals), and 99.584% for Dataset3 (fault size (mm)); a 100% average success rate is observed for Dataset4 (fault type - error free bearing (EFB), inner ring fault (IRF), outer ring fault (ORF), and ball fault (BF)).

In process monitoring, fault classification performance heavily relies on the labels of training data. However, the labeled data are inadequate and difficult to obtain because they require experienced human annotators. In this paper, a modified label propagation (MLP) method is proposed to propagate labels from labeled data to unlabeled data. The proposed label propagation algorithm has the following advantages: (1) It constructs a global and local consistency framework with the aid of a data graph, manifold learning, and data labels. This framework follows the assumption that data on the manifold will have similar structures, and nearby data will have similar labels. (2) Considering the inner relationship between the unlabeled data and historical data, a new definition for the initial label matrix is offered, which is significant for label propagation. (3) The new method propagates labels in a low-dimensional manifold space, which is different from most existing label propagation methods that propagate them in the original space. The results reveal that under the global and local consistency framework, soft labels of unlabeled data are given more effective predictions. With additional soft labels of unlabeled data, the MLP-based fault classification method is introduced. The simulation results obtained using a toy example demonstrate the label propagation performance of the MLP, and those obtained for the penicillin fermentation process verify the effectiveness of the MLP-based fault classification method.

Most high resistance faults in distribution network are caused by overhead lines contacting with high impedance objects. It is difficult to identify the high resistance faults with the steady-state characteristics in distribution network. In this paper, a Negative Selection Algorithm (NSA) based identification framework is proposed to detect the distribution network faults with high resistance. The Hilbert-Huang transform (HHT) analysis method is used to distinguish the faults from normal state. The sum of the first two order intrinsic mode function (IMF) components of zero sequence voltage within a cycle after fault is taken as the extracted characteristic of high resistance faults. An improved negative selection method is proposed to increase detection rate and realize the classification of abnormal states, so that normal training samples and a few fault samples can generate enough detector sets with higher coverage of non-self set area. Based on a 10 kV distribution network, the performance of the proposed identification framework is evaluated. The simulation results show that, compared with the wavelet analysis and the neural network algorithm, the proposed algorithm can effectively identify the high resistance faults in distribution network with small samples.

Detection and classification of any anomaly at its commencement are very crucial for optimal management of assets in power system grids. This paper presents a novel hybrid approach that combines S-transform (ST) and feedforward neural network (FFNN) for the detection and classification of distribution grid faults. In this proposed strategy, the measured three-phase current signals are processed through ST with a view to extracting useful statistical features. The extracted features are then fetched to FFNN in order to detect and classify different types of faults. The proposed approach is implemented in two different test distribution grids modeled and simulated in real-time digital simulator and MATLAB/SIMULINK. The obtained results justify the efficacy of the presented technique for both noise-free and noisy data. In addition, the developed technique is independent of fault resistance, inception angle, distance, and prefault loading condition. Besides, the comparative results confirm the superiority and competitiveness of the developed technique over the available techniques reported in the literature.

How to extract the fault feature and how to design the classification algorithm are the two most critical problems in power electronic circuits (PECs) fault diagnosis. Based on a kernel entropy component analysis theory, combining an extreme learning machine classification algorithm, this paper explores the feasibility of applying an ensemble approach, called KECA-ELM, to deal with the hard fault and soft fault diagnosis in a superbuck converter circuit (SCC). This approach can reduce the influence of the complex correlation between the data on the accuracy of fault classification. Furthermore, it can compress the feature dimension of the data while maintaining the feature discriminating power and reduce the computation in the classification stage. We record the signal from the output of the circuit, analyze their static and dynamic electrical performance, and then select the representative feature parameters. These feature parameters are combined into feature vectors which can reflect the health status of the PECs. Finally, simulation and physical experiment are presented in the SCC to demonstrate the fault classification ability of the proposed approach.

Regenerative braking is one of the most promising and environmentally friendly technologies used in electric and hybrid electric vehicles to improve energy efficiency and vehicle stability. This paper presents a systematic data-driven process for detecting and diagnosing faults in the regenerative braking system of hybrid electric vehicles. The diagnostic process involves signal processing and statistical techniques for feature extraction, data reduction for implementation in memory-constrained electronic control units, and variety of fault classification methodologies to isolate faults in the regenerative braking system. The results demonstrate that highly accurate fault diagnosis is possible with the classification methodologies. The process can be employed for fault analysis in a wide variety of systems, ranging from automobiles to buildings to aerospace systems.