Search

Power generation systems using photovoltaic (PV) technology have become increasingly popular due to their high production efficiency. A partial shading defect is the most common defect in this system under the process of production, diminishing both the amount and quality of energy produced. This paper proposes an Artificial Neural Network and Golden Eagle Optimization based prediction of the fault and its detection in a standalone PV system to recover the optimum performance and diagnosis of the PV system. The proposed technique combines the Artificial Neural Network (ANN) and Golden Eagle Optimization (GEO) algorithm. The major contribution of this work is to raise PV systems' performance. The result is a defect in the classification and identification of an ANN is used. The use of GEO provides an efficient optimization technique for ANN training, which reduces the training time and improves the accuracy of the model. The proposed technique is executed on the MATLAB site and contrasted with different present techniques, like genetic algorithm (GA),Elephant Herding Optimization (EHO) and Particle Swarm Optimization (PSO). The findings displays that the proposed technique is more accurate and effective than the existing methodologies for detecting and diagnosing defects in PV systems. • A hybrid technique for predict fault in standalone PV to improve the performance of PV • The proposed method is the joint execution of ANN and GEO • The major contribution of this work is to improve the performance of PV systems • The use of GEO provides an efficient optimization technique for ANN training • The proposed method is more accurate and effective than existing techniques [ABSTRACT FROM AUTHOR]

Line-to-line fault (LLF) is considered one of the most common electrical faults occurring on the array side of the Photovoltaic System (PVS). This research proposes a diagnostic method to detect and classify LLFs under low mismatch levels, and higher fault impedances of up to 45 ohm when compared to other methods that did not exceed 25 ohm. The proposed method uses robust ML classifiers, namely Quadratic Discriminate Analysis (QDA) and Feed-Forward Neural Network (FNN), to design the required models. For this purpose, seven efficient features including array 2-voltages, array 2-currents, fill factor, maximum power to irradiance ratio, and voltage-temperature product, are extracted from the I-V characteristics curve of the employed Photovoltaic (PV) array. This is done under normal and faulty conditions, with a wide range of climate conditions. The faults are first detected by a detection module, then classified according to their mismatch levels by using classification module. Each module uses two Machine Learning (ML) models which represent the QAD and FNN classifiers. The results demonstrate that the proposed method can detect the LLFs, even under critical conditions of mismatch and impedance, with an average accuracy of 99.81% and 100% for QDA and FNN respectively. In addition, it classifies the severity of these faults with an average accuracy of 99.28% and 100% for both adopted models. [ABSTRACT FROM AUTHOR]

The fault detection and classification (FDC) modeling proposed in this study is a research approach that is intended to improve the performance of plasma process models by leveraging optical emission spectroscopy (OES) data containing plasma information (PI) and enhancing model interpretability using explainable artificial intelligence (XAI) algorithms. Status variable identification data that included normal and abnormal states of bias power, pressure, SF6 gas flow, and O2 gas flow were collected during a silicon etching process with SF6, O2 gas plasma. Additional variables were derived from the OES data and included additional PI, such as O and F radicals, which were computed using actinometry, and electron temperature and electron density computed using the line ratio method. By building a high-performance FDC model and interpreting its results using XAI algorithms, we propose solutions to the limitations of the FDC model in semiconductor plasma processes. [ABSTRACT FROM AUTHOR]

Abstract Distributed energy generation increases the need for smart grid monitoring, protection, and control. Localization, classification, and fault detection are essential for addressing any problems immediately and resuming the smart grid as soon as possible. Simultaneously, the capacity to swiftly identify smart grid issues utilizing sensor data and easily accessible frequency and voltage data from PMU devices is a prerequisite of this task. Therefore, this paper proposes new methods using fuzzy logic and adaptive fuzzy neural networks as well as machine learning and meta-heuristic algorithms. First, line voltage is used by a fuzzy thresholding method to estimate when a transmission line defect would develop in less than 1.2 clock cycles. Next, features taken from frequency signals in the real-time interval are utilized to classify the type of error using machine learning systems (decision tree algorithm and random forest algorithm) optimized with wild horse meta-heuristic algorithm. To locate the precise problem location, we finally use a neural fuzzy inference system that is capable of adapting to new data. We employ a simulated power transmission system in MATLAB to test our proposed solutions. Mean square error (MSE) and confusion matrix are used to assess the efficiency of a classifier or detector. For the decision tree algorithm method, the detector attained an acceptable MSE of 2.34e−4 and accuracy of 98.1%, and for the random forest algorithm method, an acceptable MSE of 3.54e−6 and accuracy of 100%. Furthermore, the placement error was less than 153.6 m in any direction along the line.

The accurate detection and classification of faults in photovoltaic (PV) systems contribute to enhancing their performance. Bypass diode (BD) heating is a common issue afflicting field‐installed PV modules. Although different fault modes exist, they exhibit similar characteristics, which hinder their early detection and classification. Thus, a real‐time fault detection and classification method for heated BDs in PV modules using thermoelectric devices (TEs) is proposed. When PV modules experience partial shading (PS) or exhibit bypass diode faults (BDFs), the temperature of the BD rises. By attaching a TE to the BD, the heat generated due to the PS or BDF can be converted into electrical energy. Additionally, the cause of bypass heating can be classified in real time by analyzing the TE's power‐generation capacity, considering the inverter status and ambient conditions. For example, the power generated by the TE is in the ranges of 8.5–14.7 (PS) and 28–49.4 mW cm−2 (BDF) under operating conditions. Moreover, the power generated by the TE under the PS and BDF conditions is sufficiently high to drive a communication system, guaranteeing the reliability of the detection system. Therefore, the proposed bypass heating detection system can classify the fault modes of PV systems in real time without external power. [ABSTRACT FROM AUTHOR]

Adaptive protection schemes have been developed to address the problem of behavior-changing power systems integrated with inverter-based generation (IBG). This paper proposes a machine-learning-based fault detection and classification technique using a setting-group-based adaptation approach. Multigroup settings were designed depending on the types of power generation (synchronous generator, PV plant, and type-3 wind farm) connected to a transmission line in the 39-Bus New England System. For each system topology, an optimized pretrained ensemble tree classifier was used. The adaptation process has two phases: an offline learning phase to tune the classifiers and select the optimum subset of features, and an online phase where the circuit breaker (CB) status and the active output power of the generators are continuously monitored to identify the current system topology and to select the appropriate setting group. The proposed system achieved an average accuracy of 99.4%, a 99.5% average precision, a 99.9% average specificity, and a 99.4% average sensitivity of classification. The robustness analysis was conducted by applying several fault scenarios not considered during training, which include different transmission network configurations and different penetration levels of IBGs. The case of incorrect selection of the appropriate setting group resulting from selecting the wrong topology is also considered. It was noticed that the performance of developed classifiers deteriorates when the transmission network is reconfigured and the incorrect setting group is selected. [ABSTRACT FROM AUTHOR]

An intelligent protection scheme based on the single fuzzy inference system and discrete Fourier transform is proposed for the faulty phase detection and classification of faults on the mutually coupled double circuit transmission lines. This proposed protection technique uses the magnitude of the pre-processed current information of the sending end bus only. This is implemented in the MATLAB/Simulink environment on a 400 kV, 50 Hz, and 300 km double circuit power system model. The efficacy of the proposed scheme has been tested by performing a wide range of simulation studies concerning different types of faults (common short circuit faults, cross-country and evolving faults, and high impedance faults) and typical fault scenarios (current transformer saturation, noisy environment, and fault during power swing conditions) with variations in different fault parameters and operating conditions. The results presented confirm that the proposed method detects/classifies all the faults within one cycle time and is reliable with a detection/classification accuracy of 99.75%. It is immune to variations in fault parameters, and different operating conditions, and is not affected by the zero sequence mutual impedance. It does not require any training and communication link. Further, the performance is also appraised with other training-based protection schemes. [ABSTRACT FROM AUTHOR]

Unit-to-unit variation among the production chambers is a long-lasting challenge for Fault Detection and Classification (FDC) development in the semiconductor industry. Currently, various methods are applied for knowledge transfer among chambers and generalized FDC model development. However, the existing methods cannot give a quantitative or qualitative measure for cross-chamber data transferability evaluation. This research proposes a novel methodology for data transferability evaluation and important sensor screening, which can serve as a data quality evaluation tool for any FDC model. In this research, firstly, Time Series Alignment Kernel (TSAK) is incorporated into Multidomain Discriminant Analysis (MDA) algorithm to achieve sensor-based domain generalization. Then, domain-invariant features are directly extracted for sensor visualization. After that, Fisher’s criterion ratios of the labeled good wafer samples and defective ones are computed based on the domain-invariant features of each sensor to quantitatively estimate how easy it is to transfer knowledge of each sensor among chambers, i.e., data transferability evaluation. Lastly, the proposed method develops a Recursive Feature Elimination (RFE)-based sensor selection algorithm to qualitatively analyze the importance of each sensor channel and identify the critical sensor subset. In this study, validation of the proposed method is based on two open-source datasets from real production lines. [ABSTRACT FROM AUTHOR]

Multivariate batch time-series data sets within Semiconductor manufacturing processes present a difficult environment for effective Anomaly Detection (AD). The challenge is amplified by the limited availability of ground truth labelled data. In scenarios where AD is possible, black box modelling approaches constrain model interpretability. These challenges obstruct the widespread adoption of Deep Learning solutions. The objective of the study is to demonstrate an AD approach which employs 1-Dimensional Convolutional AutoEncoders (1d-CAE) and Localised Reconstruction Error (LRE) to improve AD performance and interpretability. Using LRE to identify sensors and data that result in the anomaly, the explainability of the Deep Learning solution is enhanced. The Tennessee Eastman Process (TEP) and LAM 9600 Metal Etcher datasets have been utilised to validate the proposed framework. The results show that the proposed LRE approach outperforms global reconstruction errors for similar model architectures achieving an AUC of 1.00. The proposed unsupervised learning approach with AE and LRE improves model explainability which is expected to be beneficial for deployment in semiconductor manufacturing where interpretable and trustworthy results are critical for process engineering teams. [ABSTRACT FROM AUTHOR]

The article discusses the importance of detecting faults in photovoltaic (PV) systems in order to improve efficiency, reliability, and safety using thermal images and computer vision. Topics include the use of variational autoencoders (VAEs) to artificially expand the data set and improve the classification task, use of three convolutional neural network (CNN) models for the classification of the images, and effectivenessof CNN models in detecting and classifying PV faults from thermal images.

• PV Defect classification using electroluminescence imaging and deep learning. • Proper ELPV Dataset labeling and categorization. • High classification accuracy using tranfered SqueezeNet and the GoogleNet. • Superior classification accuracy using developed time-saving light-weight CNN. The escalation of implementing photovoltaic (PV) power generation has paved the road to innovative remarkable applications. The technology of utilizing electroluminescence imaging (EL) has aided the early identification of faults and rapid classification of solar cells in PV panels. Recently, deep learning neural networks (DNNs) has been extensively utilized in the field of PV fault detection and classification. Despite of the good achievements in the field of DNN-based approaches, however, there is still a potential for further developments. This includes better data preparation, proper dataset categorization and designing of efficient light-weight DNNs. In this work, an efficient approach is proposed to be used for defect detection and malfunctions' classification in PV cells, based on utilizing EL-based imaging analysis. Here, three approaches were developed using multi-scale convolutional neural network (CNN) models, the former is based on deploying the pretrained SqueezeNet and the GoogleNet, in a transfer learning fashion, whereas the latter is a light-weight CNN approach (denoted as LwNet). The experiments were elaborated on the ELPV dataset after being properly modified and categized. Two scenarios were adopted: 4-class- and 8-class-classification procedures. Experimental validation of the developed CNNs have demonstrated very promising performances, especially when adopting the 8-class approach. An average accuracy of about 94.6%, 93.95%, and 96.2% was obtained using GoogleNet, SqueezeNet and LwNet, respectively. A privilege has been granted to LwNet over SqueezeNet and GoogleNet, in terms of classification performance and time saving efficiency. [ABSTRACT FROM AUTHOR]

Adaptive protection schemes have been developed to address the problem of behavior-changing power systems integrated with inverter-based generation (IBG). This paper proposes a machine-learning-based fault detection and classification technique using a setting-group-based adaptation approach. Multigroup settings were designed depending on the types of power generation (synchronous generator, PV plant, and type-3 wind farm) connected to a transmission line in the 39-Bus New England System. For each system topology, an optimized pretrained ensemble tree classifier was used. The adaptation process has two phases: an offline learning phase to tune the classifiers and select the optimum subset of features, and an online phase where the circuit breaker (CB) status and the active output power of the generators are continuously monitored to identify the current system topology and to select the appropriate setting group. The proposed system achieved an average accuracy of 99.4%, a 99.5% average precision, a 99.9% average specificity, and a 99.4% average sensitivity of classification. The robustness analysis was conducted by applying several fault scenarios not considered during training, which include different transmission network configurations and different penetration levels of IBGs. The case of incorrect selection of the appropriate setting group resulting from selecting the wrong topology is also considered. It was noticed that the performance of developed classifiers deteriorates when the transmission network is reconfigured and the incorrect setting group is selected.

Effective fault detection and classification play essential roles in reducing the hazards such as electric shocks and fire in photovoltaic (PV) systems. However, the issues of interest in fault detection and classification for PV systems remain an open-ended challenge due to manual and time-consuming processes that require the relevant domain knowledge and experience of fault diagnoses. This paper proposes a hybrid deep-learning (DL) model-based combined architectures as the novel DL approaches to achieve the real-time automatic fault detection and classification of a PV system. This research employed the wavelet packet transform (WPT) as a data preprocessing technique to handle the PV voltage signal collected and feeding them as the inputs for combined DL architectures that consist of the equilibrium optimizer algorithm (EOA) and long short-term memory (LSTM-SAE) approaches. The combined DL architectures are able to extract the fault features automatically from the preprocessed data without requiring any previous knowledge, therefore can override the traditional shortages of manual feature extraction and manual selection of optimal features from the extracted fault features. These desirable features are anticipated to speed up the fault detection and classification capability of the proposed DL model with higher accuracy. In order to determine the performance of the proposed fault model, we carried out a comprehensive evaluation study on a 250-kW grid-connected PV system. In this paper, symmetrical and asymmetrical faults have been studied involving all the phases and ground faults such as single phase to ground, phases to phase, phase to phase to ground, and three-phase to ground. The simulation results validate the efficacy of the proposed model in terms of computation time, accuracy of fault detection, and noise robustness. Comprehensive comparisons between the simulation results and previous studies demonstrate the multidisciplinary applications of the present study.

The changing landscape of power grids with distributed energy sources and power electronic devices has led to increasing relay maloperations. Wide area backup protection is necessary for the resolution of faults and for a reliable power grid. This paper presents detecting and classifying faults in transmission lines for wide-area backup protection using phasor measurement units (PMU) data. The faults are detected and classified using a Weighted Extreme Learning Machine (WELM) algorithm, which considers the variable distribution of data among the different classes using a weighted approach. The PMU signal data used were generated by the simulation of an IEEE 39 bus test system in the PowerWorld/OpenPDC/MATLAB environment. For classification, the input features data were derived using a wavelet transform-based ensemble feature extraction technique, and the WELM classifier was optimized using Particle Swarm Optimization (PSO). The PSO optimized WELM (PSO-WELM) model trained on PMU data detected faults with 100% accuracy and classified them into different types with an accuracy of 99.85%. It is validated that the PSO-WELM outperforms other known classifiers on performance comparison. The model also classified noisy data with a signal-to-noise ratio (SNR) as low as 10 dB and with an accuracy of 97%.

Faults arising in photovoltaic (PV) systems can result in major energy loss, system shutdowns, financial loss and safety breaches. It is thus crucial to detect and identify faults to improve the efficiency, reliability, and safety of PV systems. The detection of faults in large PV installations can be a tedious and timeconsuming undertaking, particularly in large-scale installations. This detection and classification of faults can be achieved using thermal images; use of computer vision can simplify and speed up the fault detection and classification process. However, a challenge often faced in computer vision tasks is the lack of sufficient data to train these models effectively. We propose the use of variational autoencoders (VAEs) as a method to artificially expand the data set in order to improve the classification task in this context. Three convolutional neural network (CNN) architectures – InceptionV3, ResNet50 and Xception – were used for the classification of the images. Our results provide evidence that CNN models can effectively detect and classify PV faults from thermal images and that VAEs provide a viable option in this application, to improve model accuracy when training data are limited. Significance: • Faults in PV systems can be labour and time consuming to detect and classify. This process can be automated by using computer vision and thermal images. • CNN models (InceptionV3, ResNet50 and Xception) are effective in the detection and classification of PV faults from thermal images. • Small data sets are a common barrier to entry for computer vision assessments. VAEs provide an effective method to artificially expand a limited data set to allow for the successful use of CNN models. • The expansion of training data using VAEs for CNN models can improve the prediction accuracy in these models.

In this study, the derivation of the Discriminative Common Vector (DCV) approach which is first introduced for a face recognition task in the insufficient data case, for the sufficient data case is obtained and it is applied for a photovoltaic (PV) panel fault detection and classification. Two experimental studies are performed including two different fault configurations. In the first experimental study, as the faulty conditions open-circuit, short-circuit, and partial shading conditions are taken and healthy condition is taken as reference. Thus, a four-class fault detection and classification scheme is constructed. In the second experimental study, the serial resistance degradation fault is considered. This fault detection and classification scheme includes four classes that are healthy and three different serial resistance degradation. The data used in the experimental studies are formed to be 1x3 dimensional vectors which include the current, voltage, and power values obtained from the simulations in the PSIM program. In all two experimental studies for each class, a discriminative common vector (DCV) which represents the common properties of that class, thus, having a high discriminative ability is obtained. As a contribution to the literature, the derivation of DCVA which has high discrimination ability for sufficient data case, and usage of it for PV panels fault detection and classification is proposed for the first time in this study. The proposed method's performance is evaluated with the performance of PCA method that is recently used for the fault detection and classification problem in PV panel systems in the literature. In the first experimental study, the proposed method's performance (99%) is obtained significantly higher than the performance of the PCA method (95%). And in the second experimental study, while PCA can only detect the faulty condition but cannot classify the serial resistance degradation, the proposed method can both detect and classify with 99% accuracy the PV panel serial resistance degradation.

Detection and classification of faults in photovoltaic (PV) module cells have become a very important issue for the efficient and reliable operation of solar power plants. In this study, an efficient convolutional neural network (CNN) model is proposed for fast and accurate detection and classification of faults in PV module cells. The proposed model is developed using the transfer learning approach with SqueezeNet, which has fewer parameters and model sizes. In order to improve the training convergence and increase the classification performance, the activation functions of the proposed model are changed and skip connections are added from the fire modules. A dataset obtained from electroluminescence (EL) images is used in the experiments. Data augmentation techniques are also applied to eliminate the imbalance of class distribution and increase the class samples. The performance of the proposed method is compared with pre-trained CNN architectures such as AlexNet, ShuffleNet, GoogLeNet, and SqueezeNet. In the experimental studies, the accuracy, precision, sensitivity, specificity, and F1-score values of the proposed method are obtained as 91.29%, 84.21%, 89.72%, 92.04%, and 86.88%, respectively. In addition, the proposed method improves the accuracy values of the other methods between 0.99% and 6.29%. When all the obtained results are analyzed, it is observed that the proposed method has an effective performance in the detection of faults in PV module cells. [ABSTRACT FROM AUTHOR]

Hardware-in-the-Loop (HIL) has been recommended by ISO 26262 as an essential test bench for determining the safety and reliability characteristics of automotive software systems (ASSs). However, due to the complexity and the huge amount of data recorded by the HIL platform during the testing process, the conventional data analysis methods used for detecting and classifying faults based on the human expert are not realizable. Therefore, the development of effective means based on the historical data set is required to analyze the records of the testing process in an efficient manner. Even though data-driven fault diagnosis is superior to other approaches, selecting the appropriate technique from the wide range of Deep Learning (DL) techniques is challenging. Moreover, the training data containing the automotive faults are rare and considered highly confidential by the automotive industry. Using hybrid DL techniques, this study proposes a novel intelligent fault detection and classification (FDC) model to be utilized during the V-cycle development process, i.e., the system integration testing phase. To this end, an HIL-based real-time fault injection framework is used to generate faulty data without altering the original system model. In addition, a combination of the Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) is employed to build the model structure. In this study, eight types of sensor faults are considered to cover the most common potential faults in the signals of ASSs. As a case study, a gasoline engine system model is used to demonstrate the capabilities and advantages of the proposed method and to verify the performance of the model. The results prove that the proposed method shows better detection and classification performance compared to other standalone DL methods. Specifically, the overall detection accuracies of the proposed structure in terms of precision, recall and F1-score are 98.86%, 98.90% and 98.88%, respectively. For classification, the experimental results also demonstrate the superiority under unseen test data with an average accuracy of 98.8%. [ABSTRACT FROM AUTHOR]

Copyright of Electric Power Automation Equipment / Dianli Zidonghua Shebei is the property of Electric Power Automation Equipment Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Fault detection and classification has been employed to enhance yield and product quality for smart semiconductor manufacturing. For early detection of abnormal events that cause defects, the status variables identification data collected by the sensors embedded in advanced machines can be analyzed to derive the actions for advanced process control and advanced equipment control. However, the validity and effectiveness of fault detection and classification technologies may highly depend on domain knowledge and experience of the process engineers who should redefine the monitoring rules quickly when new process excursion occurred especially when ramping up new technologies and products. Motivated by realistic needs, this study aims to propose a novel strategy to empower intelligent fault detection and classification that employed convolutional neural network to analyze the feature SVID data and determine the conditions of the wafers, while shorten the cycle time for self-learning from domain knowledge and redefining new monitoring rules for fault classification in real time. This approach is validated with an empirical study in a leading semiconductor manufacturing company in Taiwan. The results have demonstrated the practical viability of the developed solution. [ABSTRACT FROM AUTHOR]

Data-driven Fault Detection and Classification approaches are becoming increasingly important in semiconductor manufacturing and in other industries aiming at implementing the Zero-defect paradigm. Two of the main challenges in developing such solutions are: (i) the complexity of sensor data, that typically presents themselves in the form of time-series, requiring the employment of time-consuming and possibly sub-optimal feature extraction approaches; (ii) the fact that faults/defects may be caused by more than a single process, but in many cases they are generated by a cascade of processes. In this paper, we tackle the first issue, by considering a two-stage case study consisting of a deposition process and a rapid thermal process. The proposed approach is based on convolutional deep autoencoders employed to perform feature extraction from time-series sensor data in frontend production equipment. We will show on the reported case study, how the proposed approach outperfoms key numbers-based approaches typically used in the industry. To allow reproducibility of the reported results and to foster research in the field, we publicly share the data used in this work. [ABSTRACT FROM AUTHOR]

Wireless sensor networks (WSNs) are resource-constrained, self-organizing systems that can operate in favorable as well as hostile environments. The deployment capabilities of WSNs, especially in inaccessible and hostile environments, have increased the confidence in the use of WSNs manifold. However, data delivery in sensor networks remains intrinsically faulty and unpredictable. Transmission of erroneous data in the network wastes limited precious resources of the network (such as nodes energy, bandwidth, etc.) and reduces the fidelity of the data. Timely identification of faulty nodes is essential to minimize their consequences, enhance data quality, improve decision making, and extend the lifetime of the network. This work proposes a scheme for detecting and classifying permanent, transient, and intermittent faults in clustered WSNs. The scheme is conversation efficient and reactive. The cluster heads silently analyze information obtained from the nodes during the regular operation for identifying any suspicious behavior. This is done by utilizing moving average and correlation in the obtained data. Cluster heads initiate fault identification process only when they find any suspicious behavior. The presented model localizes certain events which attenuate over distance and neutralize the impact of distance on fault detection accuracy. Thus, our fault diagnosis and classification process utilize correlations that exist in data and handle the problem of measurement variations due to the differences in positions of sensor nodes from the event location when events attenuate over distance. NS-2 based simulation is carried out to evaluate and validate the performance of the proposed scheme. The results show a significant improvement in the performance in terms of fault detection accuracy, false alarm rate, false negative rate, fault classification accuracy and communication overhead. [ABSTRACT FROM AUTHOR]

Fault Detection and Classification (FDC) based on Machine Learning (ML) approach was used to detect and classify mount head fault in the slider attachment process which causes the machine alarm 71 to occur which leads to 2% of machine downtime. This paper has focused on the use of classified pixel surface of mount head with fault difference conditions including Healthy, Fault I, Fault II, and Fault III to detect and diagnose mount head before a vacuum leak. The Artificial Neural Network (ANN) algorithm was a proposed classification model and has to be evaluated before using in the real processes. Three features of mount head surface pixel, i.e., inner, outer, and overall areas were investigated and used as model training data set. The experiment result indicates that the classification using the ANN model with three features performed with an accuracy of 94.3%. According to the result, it was found that the reliability of the production processes of FDC technique has increased as a result of the reduction of machine downtime by 1.886%. [ABSTRACT FROM AUTHOR]

A new current injection technique and traveling‐waves‐based fast and accurate method for fault detection, location, and classification on interconnected compressed air energy storages based micro‐grid clusters presented in this paper. The compressed air energy storages based micro‐grid clusters increase due to some issues such as DC sources and loads expanding, energy‐saving, and the power quality developing. It is necessary to recognize the fault type and location to continue service and prevent outages expansion. In this method, a circuit kit joined to the network. Fault detection is done by analyzing the coupled kits currents and the traveling‐waves of the fault current and applying them to a mathematical morphology filter, in the Fault time. Determine the fault location and type done by using a mathematical morphology filter, circuit equations, and current calculations. DC pole to ground and pole to pole faults in lines, loads, and resources are considered as MTDC microgrid disturbances. The presented method was tested in a DC interconnected micro‐grid clusters connected to several energy storages and renewable resources with many faults. The results represent the validity of the proposed method. The main advantages of the proposed fault location and classification technique are higher speed and accuracy than conventional approaches. This method robustly works to variations in fault resistance, sampling frequency, and operates very fine in high impedance fault conditions. [ABSTRACT FROM AUTHOR]

Line–line (LL) and line-ground (LG) faults may not be detected by common protection devices in Photovoltaic (PV) arrays as these faults are not detectable under high fault impedance and low mismatch level. In recent years, many efforts have been devoted to overcome these challenges using intelligent methods. However, these methods could not classify the type of faults and diagnose their severity. In this article, we propose a novel and intelligent fault monitoring method to detect and classify LL and LG faults at the dc side of PV systems. For this purpose, the main features of current–voltage curves under different fault events and normal conditions are extracted. The faults are categorized using the hierarchical classification platform. Later, the LL and LG faults are detected and classified by machine learning methods. The proposed method aims to reduce the amount of dataset, which is required for the learning process, and also obtains a higher accuracy in detecting and classifying the fault events at low mismatch levels and high fault impedance compared with other fault diagnostic methods. The experimental results verify that the proposed method precisely detects and classifies LL and LG faults on PV systems under different conditions and severity with the accuracy of 96.66% and 91.66%, respectively. [ABSTRACT FROM AUTHOR]

The aim of this study was to examine the health diagnosis classification method of quadcopter structures with different mixed faults. The loosening of the motor mount, damage to the propeller, and the loosening of the arm mount were the main fault conditions investigated. Data were first acquired under non-fault conditions and the conditions of the three types of fault. Then, the features of the vibration and pulse width modulation signals were extracted by root mean square, standard deviation, and sample entropy. Moreover, the features of the audio signal were extracted by wavelet scattering, which contains time-frequency domain information that provides significant power for classification. In this paper, we propose a simple machine learning method, based on the k-Nearest Neighbor (kNN), not only for classification but also demonstrating the efficacy of the features. To test the limits of accuracy, different configurations of kNN parameters are deployed, in addition to the features. In summary, as a result of the highly efficacious features, despite mixed fault conditions, the accuracy reached 90.73%. This method improves the accuracy of mixed faults' classification and maintains a certain level of classification effectiveness. [ABSTRACT FROM AUTHOR]

In this paper, a new fast and accurate method for fault detection, location, and classification on multi-terminal DC (MTDC) distribution networks connected to renewable energy and energy storages presented. MTDC networks develop due to some issues such as DC resources and loads expanding, and try to the power quality increasing. It is important to recognize the fault type and location in order to continue service and prevent further damages. In this method, a circuit kit is connected to the network. Fault detection is performed with the measurement of the current of the connected kits and the traveling-waves of the derivative of the fault current and applying to a mathematical morphology filter, in the Fault time. The type and location of faults determinate using circuit equations and current calculations. DC series and ground arc faults are considered as DC distribution network disturbances. The presented method was tested in an MTDC network with many faults. The results illustrate the validity of the proposed method. The main advantages of the proposed fault location and classification strategy are higher accuracy and speed than conventional approaches. This method robustly operates to changing in sampling frequency, fault resistance, and works very well in high impedance fault.

With increasing the installation of the photovoltaic modules, the different failures on components of the system are dramatically increased and thus lead to a direct impact on system productivity and efficiency. Fault detection and diagnosis accurately have an extreme impact on the maintenance and reliability of photovoltaic array. Moreover, for detecting the different faults, selecting the proper indicators for monitoring the system improves the fault diagnosis techniques performance and avoids the complexity of the system. In this paper, an efficient detection and diagnosis model for different fault types is proposed. This model has sequential steps. Firstly, the performance of seven indicators is initially analyzed to predict accurately the nonlinear output behavior of the photovoltaic system under changing environmental conditions, hence select the minimum indicators to detect the typical faults. Secondly, ten fault cases, considering single-fault types and another three faults considering multi-fault types, are investigated. At the same time, the impact of these faults on the selected indicators is deeply analyzed. Finally, the typical fault types are classified and detected effectively depending on three sequential probabilistic neural network models, which give a precise classification of the data inputs. Both theoretical and experimental tests are operated to validate the performance and effectiveness of the proposed model.

This paper provides a systematic, comprehensive, up-to-date study of the various artificial intelligence (AI) techniques on the detection and classification of faults on power transmission line. We review the latest start of the art of various intelligent approaches and even discuss the differences and outcomes of implementing intelligent methods in the protection scheme and the integrated protection scheme. Besides, there has been an increase in demand and interest for the application of AI approaches in drone to aid in detection and classification of faults. However, there are not many surveys pertaining to the development of the unmanned aerial vehicle (UAV) for the application of intelligent methods in this field. Thus, we also include in this paper the literature relevant to the implementation of various intelligent approaches in the unmanned aerial vehicle (UAV) for the fault detection and classification on power transmission line. Finally, we discuss the challenges and limitations faced in the implementation and propose ways to bridge the gap in the field for further research. This comprehensive study can act as a platform for new researchers to assess the possible different intelligent methods in detecting and classifying faults on power transmission line with a set of references that were dedicated to the attentive contributions. [ABSTRACT FROM AUTHOR]

Hardware-in-the-Loop (HIL) has been recommended by ISO 26262 as an essential test bench for determining the safety and reliability characteristics of automotive software systems (ASSs). However, due to the complexity and the huge amount of data recorded by the HIL platform during the testing process, the conventional data analysis methods used for detecting and classifying faults based on the human expert are not realizable. Therefore, the development of effective means based on the historical data set is required to analyze the records of the testing process in an efficient manner. Even though data-driven fault diagnosis is superior to other approaches, selecting the appropriate technique from the wide range of Deep Learning (DL) techniques is challenging. Moreover, the training data containing the automotive faults are rare and considered highly confidential by the automotive industry. Using hybrid DL techniques, this study proposes a novel intelligent fault detection and classification (FDC) model to be utilized during the V-cycle development process, i.e., the system integration testing phase. To this end, an HIL-based real-time fault injection framework is used to generate faulty data without altering the original system model. In addition, a combination of the Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) is employed to build the model structure. In this study, eight types of sensor faults are considered to cover the most common potential faults in the signals of ASSs. As a case study, a gasoline engine system model is used to demonstrate the capabilities and advantages of the proposed method and to verify the performance of the model. The results prove that the proposed method shows better detection and classification performance compared to other standalone DL methods. Specifically, the overall detection accuracies of the proposed structure in terms of precision, recall and F1-score are 98.86%, 98.90% and 98.88%, respectively. For classification, the experimental results also demonstrate the superiority under unseen test data with an average accuracy of 98.8%.

The custom power devices play important role for enhancing the power transfer capacity of transmission system. However, these devices introduce challenges of under reach or over reach, in the protection of transmission system. This article introduces a novel, protection algorithm based on wavelet-alienation-neural technique for STATCOM-compensated transmission system. For detecting and classifying faults, approximate coefficients are computed from the postfault quarter cycle current waveforms. Fault index, which is summation of alienation coefficients (computed by approximate coefficients) of both the buses, is computed and compared with the threshold magnitude for detecting and classifying the different faults. For the determination of fault location, artificial neural network is applied, with input as three-phase approximate coefficients, evaluated from the voltage and current signals over a time duration of a quarter cycle. Robustness of the developed scheme has been validated for various faults at different locations with varying fault impedances and angles of fault incidence. [ABSTRACT FROM AUTHOR]

Feature design and selection is challenging because of huge data volume and high-mix production systems. Most engineers still rely on human experts to suggest the specific sensor channel and specific time frames of data from which to design the features. This study proposes a novel approach for important sensor screening to prioritize the useful sensor channels for FDC model development in semiconductor manufacturing. The proposed method can be used as a pre-processing step prior to feature extraction, and the selected sensor channels can be leveraged by process engineers for finer feature design. In this research, firstly, time series alignment kernels (TSAKs) are proposed to handle multivariate trace data. Then, the proposed method combines 5 different time series alignment kernels (TSAKs) with a feature selection algorithm, minimum Redundancy Maximum Relevance (mRMR), to identify the important sensor channels. Furthermore, a TSAK+Kernel Principal Component Analysis (KPCA) algorithm is proposed for a visualization tool. Lastly, the TSAK+Support Vector Machine (SVM) is employed for results validation. In this study, validation of the proposed method is based on both open-source datasets and the proprietary datasets from a real production line. [ABSTRACT FROM AUTHOR]

• A comprehensive review of LL fault diagnosis methods in PV systems. • An ensemble learning model based on probabilistic strategy is developed. • The main features of the fault are extracted from I-V curves using a Simulink based model of PV arrays. • A feature selection algorithm is used for each learning algorithm to reduce the dataset and increase the classification accuracy. • The proposed model is very reliable and able to detect and classify all the LL faults accurately. The fault diagnosis of photovoltaic (PV) arrays aims to increase the reliability and service life of PV systems. Line-Line (LL) faults may remain undetected under low mismatch level and high impedance due to low currents of faults, resulting in power losses and fire potential disaster. This paper proposes a novel and intelligent fault diagnosis method based on an ensemble learning model and Current-Voltage (I-V) characteristics to detect and classify LL faults at the DC side of PV systems. For this purpose, first, the key features are extracted via analyzing I-V characteristics under various LL fault events and normal operation. Second, a feature selection algorithm has been applied to select the best features for each learning algorithm in order to reduce the amount of data required for the learning process. Third, an ensemble learning model is developed that combines several learning algorithms based on the probabilistic strategy to achieve superior diagnostic performance. Here, we find an excellent agreement between simulation and experimental results that the proposed method can obtain higher accuracy in detecting and classifying the LL faults, even under low mismatch levels and high fault impedances. In addition, the comparison results demonstrate that the performance of the proposed method is better than individual machine learning algorithms, so that the proposed method precisely detects and classifies LL faults on PV systems under the different conditions with an average accuracy of 99% and 99.5%, respectively. [ABSTRACT FROM AUTHOR]

In this study, a new approach for time series based condition monitoring and fault diagnosis based on bidirectional recurrent neural networks is presented. The application of bidirectional recurrent neural networks essentially provide a viewpoint change on the fault diagnosis task, which allows to handle fault relations over longer time horizons helping in avoiding critical process breakdowns and increasing the overall productivity of the system. To further enhance the capability, we propose a novel procedure of data preprocessing and restructuring which enforces the generalization and a more efficient data utilization and consequently yields more efficient network training, especially for sequential fault classification task. The proposed Bidirectional Long Short Term Memory network outperforms standard recurrent architectures including vanilla recurrent neural networks, Long Short Term Memories and Gated Recurrent Units. We apply the proposed approach to the Tennessee Eastman benchmark process to test the effectiveness of the mentioned deep architectures and provide a detailed comparative analysis. The experimental results for binary as well as multi-class classification show the superior average fault detection capability of the bidirectional Long Short Term Memory Networks compared to the other architectures and to results from other state-of-the-art architectures found in the literature. • Fault detection and classification using bidirectional long-short term memory networks. • The viewpoint change of the bidirectional long-short term memory networks allows for capturing fault relationships among different process parameters over longer time horizon. • The proposed data-processing and restructuring facilitates the information storage by the long-short term memory cells with a many to one architecture for sequential fault detection. [ABSTRACT FROM AUTHOR]

The aim of this study was to examine the health diagnosis classification method of quadcopter structures with different mixed faults. The loosening of the motor mount, damage to the propeller, and the loosening of the arm mount were the main fault conditions investigated. Data were first acquired under non-fault conditions and the conditions of the three types of fault. Then, the features of the vibration and pulse width modulation signals were extracted by root mean square, standard deviation, and sample entropy. Moreover, the features of the audio signal were extracted by wavelet scattering, which contains time-frequency domain information that provides significant power for classification. In this paper, we propose a simple machine learning method, based on the k-Nearest Neighbor (kNN), not only for classification but also demonstrating the efficacy of the features. To test the limits of accuracy, different configurations of kNN parameters are deployed, in addition to the features. In summary, as a result of the highly efficacious features, despite mixed fault conditions, the accuracy reached 90.73%. This method improves the accuracy of mixed faults’ classification and maintains a certain level of classification effectiveness.

A transmission line is the main commodity of power transmission network through which power is transmitted to the utility. These lines are often swayed by accidental breakdowns owing to different random origins. Hence, researchers try to detect and track down these failures at the earliest to avoid financial prejudice. This paper offers a new real-time mathematical morphology based approach for fault feature extraction. The morphological open-close-median filter is exploited to wrest unique fault features which are then fed as an input to support vector machine to detect and classify the short circuit faults. The acquired graphical and numerical results of the extracted fault features affirm the potency of the offered scheme. The proposed scheme has been 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 in 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 complexity, in less than quarter of a cycle. [ABSTRACT FROM AUTHOR]

In this paper, a new fast and accurate method for fault detection, location, and classification on multi-terminal DC (MTDC) distribution networks connected to renewable energy and energy storages presented. MTDC networks develop due to some issues such as DC resources and loads expanding, and try to the power quality increasing. It is important to recognize the fault type and location in order to continue service and prevent further damages. In this method, a circuit kit is connected to the network. Fault detection is performed with the measurement of the current of the connected kits and the travelingwaves of the derivative of the fault current and applying to a mathematical morphology filter, in the Fault time. The type and location of faults determinate using circuit equations and current calculations. DC series and ground arc faults are considered as DC distribution network disturbances. The presented method was tested in an MTDC network with many faults. The results illustrate the validity of the proposed method. The main advantages of the proposed fault location and classification strategy are higher accuracy and speed than conventional approaches. This method robustly operates to changing in sampling frequency, fault resistance, and works very well in high impedance fault. [ABSTRACT FROM AUTHOR]

Health monitoring and fault detection are becoming more and more important in electrical machine systems due to the increasing demand for reliability. Winding turn fault is a common fault in permanent magnet (PM) machines, which can cause severe damages and requires prompt detection and mitigation. High resistance connection fault, which results in phase asymmetry may also occur but does not require immediate shutdown. Thus, apart from the fault detection, the classification between the two faults is also required. In this paper, a new technique for detecting and classifying turn fault and HRC fault by utilizing both the high and low frequency components of the zero sequence voltage is proposed with the enhanced sensitivity. The dependence on the operating conditions is minimized with the proposed fault indicators. The effectiveness of fault detection and classification has been verified by extensive experimental tests on a triple redundant fault tolerant PM assisted synchronous reluctance machine. The robustness of the turn fault detection in transient states and under no load conditions has also been demonstrated. [ABSTRACT FROM AUTHOR]

In this paper, Hilbert Huang transform based technique is introduced to detect and classify double line to ground faults in a 138 kV, 60 Hz, 68 km long six phase transmission line. Hilbert Huang transform is used to extract the hidden features contained in fault current signal in the form of Hilbert coefficients. The proposed technique does not requires the communication link between the two terminals of six phase transmission line since the fault current signals are recorded at the first terminal only i.e. at bus-1. For the justification of the proposed technique, wide variety of fault test studies were carried out in MATLAB/ Simscape power system toolbox for different types of three phase to ground faults by varying fault type, fault location, fault inception time, fault resistance and ground resistance. Test results shows that the Hilbert Huang transform based proposed technique effectively detects/ classifies the double line to ground faults.

The detection of wafer faults in early process steps through monitoring and analyzing multivariate process trace data contribute to wafer yield improvements. Standard classification algorithms have been generally used for fault detection and classification (FDC). However, this approach can cause information loss while extracting statistical features from the trace data and cannot consider class imbalance situations where much fewer faulty wafers are generated than normal wafers. In addition, the approach does not consider normal wafer-to-wafer (W2W) variations and sensor noise inherent in the trace data. These drawbacks significantly degrade FDC performance. This paper proposes a method that builds an FDC model only with trace data of normal wafers in which W2W variations and sensor noise exist. The one-class FDC method detects the occurrence of abnormal trace patterns that cause wafer faults by removing W2W variations and sensor noise from raw traces by using denoising autoencoders, and this method finds the fault-introducing process parameters with the occurrence times. In experiments using the trace data of etch and chemical vapor deposition processes, the proposed method exhibited 1% and 6% higher performance than the best-performing method among comparison methods in terms of the geometric mean of the normal and fault detection accuracies. [ABSTRACT FROM AUTHOR]