Efficient Fault Detection Research Articles

Gear Fault Diagnosis Based on Variational Modal Decomposition and Wide+Narrow Visual Field Neural Networks

In modern industrial production, rotating machinery plays an important role. The gears in this machinery adjust the speed and transmission of torque. Therefore, when the gear fails, it is very important to be able to diagnose the fault quickly and accurately. Gear vibration signals are often used in gear fault diagnosis, but the fault signal is often overwhelmed by noises. To enable the scientific and efficient detection of faults, this study proposes a gear fault diagnosis method based on variational modal decomposition (VMD) and wide+narrow visual field neural networks (WNVNNs), namely VMD-WNVNN. VMD-WNVNN consists of two stages. In the feature extraction stage, VMD and Pearson correlation coefficients are used to decompose and reconstruct the original data to obtain the features of these data in the frequency domain. In the classification stage, WNVNN is used to classify the data based on features. The final results of the gear fault diagnosis experiments show that this method not only has higher classification accuracy but also has higher classification stability than other recently proposed methods. Note to Practitioners—The gearbox is composed of many mechanical parts, such as gears, shafts, and bearings. Therefore, the vibration signal collected by the vibration sensor on the gearbox housing will contain the vibration signal of each part and the noise caused by processing errors. Therefore, if some methods can be used to efficiently extract the characteristic signals required for diagnosis in the data processing stage, the efficiency of fault diagnosis will be greatly improved. This article takes the health of gears as the research object and proposes a method that combines adaptive signal decomposition and deep learning technology. Experimental results show that this method has higher classification accuracy and classification stability than other methods proposed recently.

PV System Failures Diagnosis Based on Multiscale Dispersion Entropy.

Photovoltaic (PV) system diagnosis is a growing research domain likewise solar energy’s ongoing significant expansion. Indeed, efficient Fault Detection and Diagnosis (FDD) tools are crucial to guarantee reliability, avoid premature aging and improve the profitability of PV plants. In this paper, an on-line diagnosis method using the PV plant electrical output is presented. This entirely signal-based method combines variational mode decomposition (VMD) and multiscale dispersion entropy (MDE) for the purpose of detecting and isolating faults in a real grid-connected PV plant. The present method seeks a low-cost design, an ease of implementation and a low computation cost. Taking into account the innovation of applying these techniques to PV FDD, the VMD and MDE procedures as well as parameters identification are carefully detailed. The proposed FFD approach performance is assessed on a real rooftop PV plant with experimentally induced faults, and the first results reveal the MDE approach has good suitability for PV plants diagnosis.

Infrared thermography enabled morphology detection approach for additive manufactured flexible electrodes

Additive manufactured flexible electrodes show prospect applications in flexible electronics. However, morphological defects may affect a flexible electronic device's sensitivity, response value, reliability, etc. Therefore, accurate and efficient defect detection is still challenging, especially for mass-manufactured flexible electrodes. This paper proposed and tested a rapid morphological defects detection approach with infrared thermography technology. This approach shows advantages of high detect resolution (minimum test defect: 183.38um), good linearity (R2 = 0.9887), efficiency, and nondestructive and large-scale applications, thereby achieving quality control and process evaluation in the mass additive manufacturing of flexible electrodes. Furthermore, resulting in improves the reliability and consistency of flexible electronic devices.

Self-healing and optimal fault tolerant routing in wireless sensor networks using genetical swarm optimization

A wireless sensor network (WSN) is used in area monitoring, surveillance, virtual reality, artificial intelligence, etc. The interaction between sensor nodes (SNs) and base station (BS) play a major role in WSN. SNs use more energy to transmit data, and the occurrence of faults in these nodes may lead to complete network failure. To ensure enhanced network performance, the incorporation of an efficient fault detection mechanism becomes essential in WSN. The major aim of the work is to improve the self-healing capacity of the network. To meet the above requirement, a fault-tolerant routing path identification with genetical swarm optimization (FTGSO) is introduced in this work. Genetical swarm optimization (GSO) is the integration of the two optimizations like Genetic Algorithm (GA) and Particle swarm optimization (PSO). The cluster head (CH) selection carried out on the basis of residual energy, coverage, communication cost and proximity. Further, a fault-free routing path is introduced by GSO, and a self-healing method is employed to resolve any network connectivity issues and resume the normal system operation. The analysis and simulation outcomes are compared with other optimization algorithms to verify the effectiveness. The evaluation results verify the enhanced performance of the introduced scheme with a higher PDR (packet delivery ratio) of 96.8%, lower energy consumption of 0.19 J with a minimum packet loss ratio of 3.2% and 30 ms of E2E (end-to-end delay) compared to other existing routing protocols. Also, the performance of the proposed method is compared on the basis of E2E, energy efficiency, (PLR) packet loss ratio and PDR.

An Efficient Fault Detection Method for Induction Motors Using Thermal Imaging and Machine Vision

Induction motors (IMs) are the backbone of industry, and play a vital role in daily life as well. However, induction motors face various faults during their operation, which may cause overheating, energy losses, and failure in the motors. Keeping in mind the severity of the issues associated with fault occurrence, this paper proposes a novel method of fault detection in induction motors by using “Machine Vision (MV)” along with “Infrared Thermography (IRT)”. It is worth mentioning that the timely prevention of faults in the IM ensures the motor’s safety from failures, and provides longer service life. In this work, a dataset of thermal images of an induction motor under different conditions (i.e., normal operation, overloaded, and fault) was developed using an infrared camera without disturbing the working condition of the motor. Then, the extracted thermal images were effectively used for the feature extraction and training by local octa pattern (LOP) and support-vector machine (SVM) classifiers, respectively. In order to enhance the quality of feature extraction from images, the LOP was implemented along with a genetic algorithm (GA). Finally, the proposed methodology was implemented and validated by detecting the faults introduced in an induction motor in real time. In addition to that, a comparative study of the suggested methodology with existing methods also verified the supremacy and effectiveness of the proposed method in comparison to the previous techniques.

A novel fault detection and identification system for solar PV systems using truncated arrangement of active cell (TAAC) structure

Fault detection and identification in a solar Photovoltaic (PV) systems are one of the crucial task in recent days for ensuring both reliability and safety measures. The fault occurrence in the PV cell will affect the output power, and can reduce the efficiency of its characteristics. The fault in PV cell can identify by using the thermal scan method manually. Arrangement of the proposed setup regularly is not possible to monitor due to the hardware installation of several equipment, it took more time to test, and validate the affected PV cells prediction less accuracy while doing in manual testing. In order to solve these issues, this paper intends to propose a novel algorithm, named as Truncated Arrangement of Active Cell (TAAC) structure for accurately detecting the PV faults. This technique is used to analyze the PV cell aging condition and to enhance the PV characteristics. Typically, the improvement in a cell arrangement provides an optimal solution for efficient fault detection. Moreover, the TAAC architecture computes the optimal solution for a PV output terminal based on the PV cell parameters and variation of temperature measures. Also, a Kalman filtering technique is employed to extract the features that are used to improve the detection process. The major advantages of this structure are, it enhance the lifetime of PV cell and stores the maximum power for a long time usage. The experimental results evaluate the performance of this technique by using various measures such as false alarm rate, misclassification rate, misdetection rate, and prediction rate. Furthermore, some of the existing techniques are compared with the proposed technique for proving its superiority.

ESDDNet: efficient small defect detection network of workpiece surface

Defect detection of a workpiece surface is a basic and essential task in the production of products. Although significant progress has been made in workpiece surface defect detection, traditional methods still find it difficult to detect small defects efficiently. To deal with this problem, we propose an efficient small defect detection network with a novel parallel convolution module, serial convolution module and feature fusion module. First, a lightweight backbone network is used to extract the preliminary defect features. Second, the parallel convolution module and serial convolution module are used to obtain the abundant defect features. Then, the feature fusion module is used to fuse the shallow features with deep features, to enhance the features of the small defects. Finally, the obtained features are put into the corresponding detection head to get the final prediction results. The experimental results on a local cable dataset and a public printed circuit board dataset show that our method achieves a remarkable performance in detecting small defects and achieves a favorable trade-off between accuracy, speed and model size, which meets the requirements of industrial applications.

A Deep Convolutional Generative Adversarial Networks-Based Method for Defect Detection in Small Sample Industrial Parts Images

Online defect detection in small industrial parts is of paramount importance for building closed loop intelligent manufacturing systems. However, high-efficiency and high-precision detection of surface defects in these manufacturing systems is a difficult task and poses a major research challenge. The small sample size of industrial parts available for training machine learning algorithms and the low accuracy of computer vision-based inspection algorithms are the bottlenecks that restrict the development of efficient online defect detection technology. To address these issues, we propose a small sample gear face defect detection method based on a Deep Convolutional Generative Adversarial Network (DCGAN) and a lightweight Convolutional Neural Network (CNN) in this paper. Initially, we perform data augmentation by using DCGAN and traditional data enhancement methods which effectively increase the size of the training data. In the next stage, we perform defect classification by using a lightweight CNN model which is based on the state-of-the-art Vgg11 network. We introduce the Leaky ReLU activation function and a dropout layer in the proposed CNN. In the experimental evaluation, the proposed framework achieves a high score of 98.40%, which is better than that of the classic Vgg11 network model. The method proposed in this paper is helpful for the detection of defects in industrial parts when the available sample size for training is small.

Online Detection of Fabric Defects Based on Improved CenterNet with Deformable Convolution.

The traditional manual defect detection method has low efficiency and is time-consuming and laborious. To address this issue, this paper proposed an automatic detection framework for fabric defect detection, which consists of a hardware system and detection algorithm. For the efficient and high-quality acquisition of fabric images, an image acquisition assembly equipped with three sets of lights sources, eight cameras, and a mirror was developed. The image acquisition speed of the developed device is up to 65 m per minute of fabric. This study treats the problem of fabric defect detection as an object detection task in machine vision. Considering the real-time and precision requirements of detection, we improved some components of CenterNet to achieve efficient fabric defect detection, including the introduction of deformable convolution to adapt to different defect shapes and the introduction of i-FPN to adapt to defects of different sizes. Ablation studies demonstrate the effectiveness of our proposed improvements. The comparative experimental results show that our method achieves a satisfactory balance of accuracy and speed, which demonstrate the superiority of the proposed method. The maximum detection speed of the developed system can reach 37.3 m per minute, which can meet the real-time requirements.

Efficient surface defect detection using self-supervised learning strategy and segmentation network

Surface defect detection plays a crucial role in the production process to ensure product quality. With the development of Industry 4.0 and smart manufacturing, traditional manual defect detection becomes no longer satisfactory, and deep learning-based technologies are gradually applied to surface defect detection tasks. However, the application of deep learning-based defect detection methods in actual production lines is often constrained by insufficient data, expensive annotations, and limited computing resources. Detection methods are expected to require fewer annotations as well as smaller computational consumption. In this paper, we propose the Self-Supervised Efficient Defect Detector (SEDD), a high-efficiency defect defector based on self-supervised learning strategy and image segmentation. The self-supervised learning strategy with homographic enhancement is employed to ensure that defective samples with annotations are no longer needed in our pipeline, while competitive performance can still be achieved. Based on this strategy, a new surface defect simulation dataset generation method is proposed to solve the problem of insufficient training data. Also, a lightweight structure with the attention module is designed to reduce the computation cost without incurring accuracy. Furthermore, a multi-task auxiliary strategy is employed to reduce segmentation errors of edges. The proposed model has been evaluated with three typical datasets and achieves competitive performance compared with other tested methods, with 98.40% AUC and 74.84% AP on average. Experimental results show that our network has the smallest computational consumption and the highest running speed among the networks tested.

A comparative study for stator winding inter-turn short-circuit fault detection based on harmonic analysis of induction machine signatures

This article deals with inter-turn short-circuit fault detection in stator windings of squirrel cage induction machines. The main aim is to perform harmonic analysis of different electrical signatures namely the stator phase current, external magnetic flux and electromagnetic torque at different levels of mechanical load in order to develop an efficient fault detection approach of this kind of defect in induction machines. The proposed approach is based on the analysis of saturation related harmonics at rank 3k1fs, where k1 is an odd number, and magnetomotive force (MMF)-related harmonics at rank (6k2±1)fs, with k2=1,2,… in stator phase current and stray flux and harmonics at rank 2k2fs in electromagnetic torque. The amplitudes of these last harmonics in healthy condition are compared with 3% power supply unbalance, 16.6% (40 turns) and 33% (80 turns) levels of inter-turn short-circuit fault in frequency range from 0 Hz to 2500 Hz under different levels of mechanical load. Besides, the stand-still test is also investigated in this work. Simulation study is carried out based on 2.2kW squirrel cage IM using finite element method (FEM). This method provides accurate and inexpensive tool for evaluating the performance of induction machine under healthy and faulty conditions. The obtained results demonstrate that the stray flux is the most sensitive signature to the stator winding inter-turn short-circuit fault, and it is robust against the power supply unbalance in comparison with both stator current and electromagnetic torque.

An improvement of both security and reliability for AES implementations

Cryptographic circuits, because they contain confidential information, are subject to fraudulent manipulations, commonly called attacks, by ill-intentioned people. Several attacks have been identified and analyzed. One of the most efficient attacks, called DFA (Differential Fault Analysis) cryptanalysis, exploits the presence of faults, injected voluntarily by the attacker, for example with a laser, in the calculations. Countermeasures are then developed and validated to protect the hardware implementation of cryptographic algorithms such as AES algorithm. The present paper aims to protect the AES algorithm against fault injection attacks. In this paper, an efficient fault detection method for the AES algorithm has been developed. In our proposed design, the AES round architecture is devised into three parts and two pipelines registers are inserted in between. Our simulations show that fault detection capabilities of the proposed method for random fault archives 99.539%. The error detection structure can detect the inserted faults with good coverage using the proposed approach. In order to verify the accuracy of our results, we compare our implementation result with the results presented in the literature. Note that the results obtained are competitive with existing implementations. Our fault detection method has the ability to achieve a compromise of safety level and low implementation cost.

Hessian Locally Linear Embedding of PMU Data for Efficient Fault Detection in Power Systems

This article develops a computationally efficient fault-detection method for power systems, by exploiting phasor measurement unit (PMU) data and Hessian locally linear embedding (HLLE) technique. First, via HLLE technique, high-dimensional PMU data are transformed into low-dimensional embedding coordinates, which effectively captures the fluctuations of PMU data. Next, based on the feature space of embedding coordinates, T-squared statistic is employed for online fault detection. The method is evaluated on IEEE 39-bus system and a real-world system, exhibiting decent fault-detection performance as well as a considerably lower computational complexity when compared with existent methods.

A Novel Eddy Current Method for Defect Detection Immune to Lift-Off

Eddy current (EC) sensors have been applied to detect defects in many years. Since both defects and the lift-off variation will lead to measurement signal fluctuation, it is necessary to distinguish the defect response and lift-off variation response. In this work, a novel eddy current method is proposed for defect detection immune to lift-off via a triple-coil structure. Firstly, a simplified theoretical model is built to investigate the relationship between the lift-off and the inductance variation. It is found the lift-off is proportional to the logarithm of inductance variation in real part under high frequency. In terms of sensor design, the triple-coil sensor consists of two pairs of transmitting-receiving coils, from which one of the receiving coils is larger than the other one. With such sensor structure, the small size coil aims to detect the defect, while the larger size coil aims to calibrate lift-off. The proposed method is verified by experiments with the sensor scanning over a Titanium specimen with defects of different sizes. The results indicate the proposed method can achieve efficient and precise defect detection immune to lift-off.

Deep Learning-Based Algorithm for Multi-Type Defects Detection in Solar Cells with Aerial EL Images for Photovoltaic Plants

Defects detection with Electroluminescence (EL) image for photovoltaic (PV) module has become a standard test procedure during the process of production, installation, and operation of solar modules. There are some typical defects types, such as crack, finger interruption, that can be recognized with high accuracy. However, due to the complexity of EL images and the limitation of the dataset, it is hard to label all types of defects during the inspection process. The unknown or unlabeled create significant difficulties in the practical application of the automatic defects detection technique. To address the problem, we proposed an evolutionary algorithm combined with traditional image processing technology, deep learning, transfer learning, and deep clustering, which can recognize the unknown or unlabeled in the original dataset defects automatically along with the increasing of the dataset size. Specifically, we first propose a deep learning-based features extractor and defects classifier. Then, the unlabeled defects can be classified by the deep clustering algorithm and stored separately to update the original database without human intervention. When the number of unknown images reaches the preset values, transfer learning is introduced to train the classifier with the updated database. The fine-tuned model can detect new defects with high accuracy. Finally, numerical results confirm that the proposed solution can carry out efficient and accurate defect detection automatically using electroluminescence images.

An Efficient AES 32-Bit Architecture Resistant to Fault Attacks

The Advanced Encryption Standard cryptographic algorithm, named AES, is implemented in cryptographic circuits to ensure high security level to any system which required confidentiality and secure information exchange. One of the most effective physical attacks against the hardware implementation of AES is fault attacks which can extract secret data. Until now, a several AES fault detection schemes against fault injection attacks have been proposed. In this paper, so as to ensure a high level of security against fault injection attacks, a new efficient fault detection scheme based on the AES architecture modification has been proposed. For this reason, the AES 32-bit round is divided into two half rounds and input and pipeline registers are implemented between them. The proposed scheme is independent of the procedure the AES is implemented. Thus, it can be implemented to secure the pipeline and iterative architectures. To evaluate the robustness of the proposed fault detection scheme against fault injection attacks, we conduct a transient and permanent fault attacks and then we determine the fault detection capability; it is about 99.88585% and 99.9069% for transient and permanent faults respectively. We have modeled the AES fault detection scheme using VHDL hardware language and through hardware FPGA implementation. The FPGA results demonstrate that our scheme can efficiently protect the AES hardware implementation against fault attacks. It can be simply implemented with low complexity. In addition, the FPGA implementation performances prove the low area overhead and the high efficiency and working frequency for the proposed AES detection scheme.

Fuzzy Logic and Modified Butterfly Optimization with Efficient Fault Detection and Recovery Mechanisms for Secured Fault-Tolerant Routing in Wireless Sensor Networks

Fuzzy Logic and Modified Butterfly Optimization with Efficient Fault Detection and Recovery Mechanisms for Secured Fault-Tolerant Routing in Wireless Sensor Networks

Improved neural network-based sensor fault detection and estimation strategy for an autonomous aerial vehicle

PurposeThis paper aims to design an adaptive nonlinear strategy capable of timely detection and reconstruction of faults in the attitude’s sensors of an autonomous aerial vehicle with greater accuracy concerning other conventional approaches in the literature.Design/methodology/approachThe proposed scheme integrates a baseline nonlinear controller with an improved radial basis function neural network (IRBFNN) to detect different kinds of anomalies and failures that may occur in the attitude’s sensors of an autonomous aerial vehicle. An integral sliding mode concept is used as auto-tune weight update law in the IRBFNN instead of conventional weight update laws to optimize its learning capability without computational complexities. The simulations results and stability analysis validate the promising contributions of the suggested methodology over the other conventional approaches.FindingsThe performance of the proposed control algorithm is compared with the conventional radial basis function neural network (RBFNN), multi-layer perceptron neural network (MLPNN) and high gain observer (HGO) for a quadrotor vehicle suffering from various kinds of faults, e.g. abrupt, incipient and intermittent. From the simulation results obtained, it is found that the proposed algorithm’s performance in faults detection and estimation is relatively better than the rest of the methodologies.Practical implicationsFor the improvement in the stability and safety of an autonomous aerial vehicle during flight operations, quick identification and reconstruction of attitude’s sensor faults and failures always play a crucial role. Efficient fault detection and estimation scheme are considered indispensable for an error-free and safe flight mission of an autonomous aerial vehicle.Originality/valueThe proposed scheme introduces RBFNN techniques to detect and estimate the quadrotor attitude’s sensor faults and failures efficiently. An integral sliding mode effect is used as the network’s backpropagation law to automatically modify its learning parameters accordingly, thereby speeding up the learning capabilities as compared to the conventional neural network backpropagation laws. Compared with the other investigated techniques, the proposed strategy achieve remarkable results in the detection and estimation of various faults.

Broken Rotor Bar Detection in Induction Motors through Contrast Estimation.

Induction motors (IM) are key components of any industrial process; hence, it is important to carry out continuous monitoring to detect incipient faults in them in order to avoid interruptions on production lines. Broken rotor bars (BRBs), which are among the most regular and most complex to detect faults, have attracted the attention of many researchers, who are searching for reliable methods to recognize this condition with high certainty. Most proposed techniques in the literature are applied during the IM startup transient, making it necessary to develop more efficient fault detection techniques able to carry out fault identification during the IM steady state. In this work, a novel methodology based on motor current signal analysis and contrast estimation is introduced for BRB detection. It is worth noting that contrast has mainly been used in image processing for analyzing texture, and, to the best of the authors’ knowledge, it has never been used for diagnosing the operative condition of an induction motor. Experimental results from applying the approach put forward validate Unser and Tamura contrast definitions as useful indicators for identifying and classifying an IM operational condition as healthy, one broken bar (1BB), or two broken bars (2BB), with high certainty during its steady state.

Real-World Data-Driven Machine-Learning-Based Optimal Sensor Selection Approach for Equipment Fault Detection in a Thermal Power Plant

Due to growing electricity demand, developing an efficient fault-detection system in thermal power plants (TPPs) has become a demanding issue. The most probable reason for failure in TPPs is equipment (boiler and turbine) fault. Advance detection of equipment fault can help secure maintenance shutdowns and enhance the capacity utilization rates of the equipment. Recently, an intelligent fault diagnosis based on multivariate algorithms has been introduced in TPPs. In TPPs, a huge number of sensors are used for process maintenance. However, not all of these sensors are sensitive to fault detection. The previous studies just relied on the experts’ provided data for equipment fault detection in TPPs. However, the performance of multivariate algorithms for fault detection is heavily dependent on the number of input sensors. The redundant and irrelevant sensors may reduce the performance of these algorithms, thus creating a need to determine the optimal sensor arrangement for efficient fault detection in TPPs. Therefore, this study proposes a novel machine-learning-based optimal sensor selection approach to analyze the boiler and turbine faults. Finally, real-world power plant equipment fault scenarios (boiler water wall tube leakage and turbine electric motor failure) are employed to verify the performance of the proposed model. The computational results indicate that the proposed approach enhanced the computational efficiency of machine-learning models by reducing the number of sensors up to 44% in the water wall tube leakage case scenario and 55% in the turbine motor fault case scenario. Further, the machine-learning performance is improved up to 97.6% and 92.6% in the water wall tube leakage and turbine motor fault case scenarios, respectively.