Fault Classification Research Articles

An enhanced meta-learning network with sensitivity penalty for cross-domain few-shot fault diagnosis

Abstract Big data-driven rotating machine intelligent diagnostic technology has gained widespread applications. In practice, however, fault data are limited as well as inconsistencies in fault categories among different domains are widespread. These make developing robust intelligent diagnostic models a challenge. To this end, this paper develops an enhanced meta-learning network with a sensitivity penalization mechanism (EMLN-SP) for few-shot fault diagnosis in severe domain bias. First, lightweight channel attention is introduced to establish an enhanced feature encoder under meta-learning framework, which elevates the key feature expression to facilitate the extraction of generalized diagnostic knowledge within limited samples. Second, a boundary-enhanced loss calculation method is designed, which boosts the focus for decision boundary information to prevent the model from the overfitting dilemma in the case of few-shot. Finally, a sensitivity penalty mechanism is constructed to adjust the optimization direction, which prevents the model from falling into a local optimum, to boost the generalization of the model performance. The effectiveness of EMLN-SP is validated by three cross-domain diagnostic cases with diverse domain offsets.

An Efficient Approach for Automatic Fault Classification Based on Data Balance and One-Dimensional Deep Learning

Predictive maintenance (PdM) is implemented to efficiently manage maintenance schedules of machinery and equipment in manufacturing by predicting potential faults with advanced technologies such as sensors, data analysis, and machine learning algorithms. This paper introduces a study of different methodologies for automatically classifying the failures in PdM data. We first present the performance evaluation of fault classification performed by shallow machine learning (SML) methods such as Decision Trees, Support Vector Machines, k-Nearest Neighbors, and one-dimensional deep learning (DL) techniques like 1D-LeNet, 1D-AlexNet, and 1D-VGG16. Then, we apply normalization, which is a scaling technique in which features are shifted and rescaled in the dataset. We reapply classification algorithms to the normalized dataset and present the performance tables in comparison with the first results we obtained. Moreover, in contrast to existing studies in the literature, we generate balanced dataset groups by randomly selecting normal data and all faulty data for all fault types from the original dataset. The dataset groups are generated with 100 different repetitions, recording performance scores for each one and presenting the maximum scores. All methods utilized in the study are similarly employed on these groups. From these scores, the use of 1D-LeNet deep learning classifiers and feature normalization resulted in achieving the highest overall accuracy and F1-score performance of 98.50% and 98.32%, respectively. As a result, the goal of this study was to develop an efficient approach for automatic fault classification, leveraging data balance, and additionally, to provide an analysis of one-dimensional deep learning and shallow machine learning-based classification methods. In light of the experimentation and comparative analysis, this study successfully achieves its stated goal by demonstrating that one-dimensional deep learning and data balance collectively emerge as the optimal approach, offering good prediction accuracy.

PCDC: Prototype-assisted dual-contrastive learning with depthwise separable convolutional neural network for few-shot fault diagnosis of permanent magnet synchronous motors under new operating conditions

Abstract The fault diagnosis of permanent magnet synchronous motor is of vital importance in industrial fields to ensure user safety and minimize economic losses from accidents. However, recent fault diagnosis methods, particularly the methods using deep learning, require a massive amount of labeled data, which may not be available in industrial fields. Few-shot learning has been recently applied in fault diagnosis for rotary machineries, to alleviate the data deficiency and/or to enable unseen fault diagnosis. However, two major obstacles still remain, specifically: a) the limited ability of the models to be generalized for use under new operating conditions and b) insufficient discriminative features to precisely diagnose fault types. To address these limitations, this study proposes a Prototype-assisted dual-Contrastive learning with Depthwise separable Convolutional neural network (PCDC) for few-shot fault diagnosis for permanent magnet synchronous motors under new working conditions. Operation-robust fault features are extracted to reinforce generalization of PCDC under new operating conditions by extracting fault-induced amplitude and frequency modulation features and by eliminating the influence of operating conditions from the motor stator current signals. Prototype-assisted dual-contrastive learning is proposed to clearly distinguish the fault categories even when the fault features are similar to each other by learning both local- and global-similarity features, which increases the instance-discrimination ability while alleviating an overfitting issue. Experimental results show that the proposed PCDC outperforms the comparison models in few-shot fault diagnosis tasks under new operating conditions.

Digital Shadow-based Control of Temperature and Fault Classification in Shell and Tube Heat Exchanger using Fuzzy Logic Technique

Digital Shadow-based Control of Temperature and Fault Classification in Shell and Tube Heat Exchanger using Fuzzy Logic Technique

Research on intelligent classification of weld defects based on improved ResNeXt network

Welding technology is widely used in the connection of metal workpieces, and then due to improper operation and other reasons, small defects may occur in the welding, which will lead to welding failure and hazardous accidents. Therefore, it is very important to accurately identify welding defects. The automatic defect identification method based on deep learning learns features directly from the full-focus mapping of weld defects and automatically classifies them; the ResNeXt network is selected for improvement, and the SK convolution unit is introduced into the original ResNeXt-50 network model to enhance the adaptive adjustment of sensory field; in order to alleviate the problem of the point-by-point convolution in the ResNeXt-50 which consumes a large amount of In order to alleviate the problem that point-by-point convolution in ResNeXt-50 consumes a large amount of computation, 1×1 group convolution is used instead of the original point-by-point convolution, and channel rearrangement is introduced to alleviate the problem of weakened information exchange of groups caused by group convolution. After improvement, the ResNeXt-50 network model can achieve a classification accuracy of 98.2%, which is 5.5% more than that of the original ResNeXt-50 on this paper's dataset, and the classification accuracy after T-sne visualization also has obvious separation boundary and clustering effects.

Model based fault detection diagnosis and classification in shell and tube heat exchanger using intelligent techniques

A method for Model-Based Approach for Fault Detection and Diagnosis has been put forth in this research. The purpose of this study is to build a heat exchanger model and to locate, categorize faults in the heat exchanger system. One of the known methods for fault control in industrial processes is fault detection and diagnosis (FDD). Both linear and non-linear systems can benefit from the FDD technique; this work describes fault identification and diagnosis for the non-linear system, the heat exchanger. A realistic mathematical model of the heat exchanger with its set of inputs and outputs has therefore been constructed, as the fault detection of the heat exchanger is related to the mathematical model for its operation. Once the mathematical model has been realized using the formulated equation, the process passes on to fault detection. Residual, which defines the failure symptoms of the system, is used to detect faults. In order to prevent failures and defective circumstances from occurring in existing systems, observer design is employed as one of the most powerful and reliable approaches residual generation. Fault diagnosis is carried out using a Fuzzy logic controller (FLC), which is designed based on the residuals obtained, and fault classification is made using the FLC via a fixed threshold. The paper's primary goals are to reduce errors and enhance non-linear systems' overall performance in industrial processes. According to the findings of the simulation, errors in sensors, processes, and actuators can be classified according to a set threshold, and they can be controlled using a traditional PID controller.

An Adaptive Image Segmentation Network for Surface Defect Detection.

Surface defect detection plays an essential role in industry, and it is challenging due to the following problems: 1) the similarity between defect and nondefect texture is very high, which eventually leads to recognition or classification errors and 2) the size of defects is tiny, which are much more difficult to be detected than larger ones. To address such problems, this article proposes an adaptive image segmentation network (AIS-Net) for pixelwise segmentation of surface defects. It consists of three main parts: multishuffle-block dilated convolution (MSDC), dual attention context guidance (DACG), and adaptive category prediction (ACP) modules, where MSDC is designed to merge the multiscale defect features for avoiding the loss of tiny defect feature caused by model depth, DACG is designed to capture more contextual information from the defect feature map for locating defect regions and obtaining clear segmentation boundaries, and ACP is used to make classification and regression for predicting defect categories. Experimental results show that the proposed AIS-Net is superior to the state-of-the-art approaches on four actual surface defect datasets (NEU-DET: 98.38% ± 0.03%, DAGM: 99.25% ± 0.02%, Magnetic-tile: 98.73% ± 0.13%, and MVTec: 99.72% ± 0.02%).

Corrugation and Squat Classification and Detection with VGG16 and YOLOv5 Neural Network Models

Railway track defects in Malaysia pose significant risks of train derailments and accidents, underscoring the urgency for early and accurate defect detection and classification. This study presents a novel approach utilizing deep learning models, VGG16 and YOLOv5, for detecting and classifying railway track defects, explicitly focusing on corrugation and squat defects. The research's uniqueness lies in its application of these specific models and the composition of a dataset collected from extensive field measurements and inspections across various railway tracks within the Track Network Maintenance Ampang Line in Malaysia. The results demonstrate that these models achieve high precision in defect classification and detection of defects by more than 80%. The proposed methodology provides the railway industry with a powerful tool to streamline maintenance planning and prioritize defect remediation efficiently. Early defect detection can prevent potential accidents and improve safety and operational efficiency. Future studies can expand on these findings by exploring the extension of the proposed techniques to address other types of rail defects. Incorporating a diverse range of scenarios and operating conditions in the dataset could further enhance the models' performance and generalization. Real-time deployment and integration with existing maintenance systems are crucial for practical adoption. This research has strengths but acknowledges limitations. Additional evaluation metrics and a diverse dataset are essential for model performance. Leveraging deep learning models offers a reliable solution for railway maintenance, enhancing safety and efficiency. Addressing these limitations will drive proactive defect management, ensuring safe and reliable railway networks.

Detection and classification of photovoltaic module defects based on artificial intelligence

Detection and classification of photovoltaic module defects based on artificial intelligence

A Deep Learning-based Fault Detection and Classification in Smart Electrical Power Transmission System

Progressively, the energy demands and responsibilities to control the demands have expanded dramatically. Subsequently, various solutions have been introduced, including producing high-capacity electrical generating power plants, and applying the grid concept to synchronize the electrical power plants in geographically scattered grids. Electrical Power Transmission Networks (EPTN) are made of many complex, dynamic, and interrelated components. The transmission lines are essential components of the EPTN, and their fundamental duty is to transport electricity from the source area to the distribution network. These components, among others, are continually prone to electrical disturbance or failure. Hence, the EPTN required fault detection and activation of protective mechanisms in the shortest time possible to preserve stability. This research focuses on using a deep learning approach for early fault detection to improve the stability of the EPTN. Early fault detection swiftly identifies and isolates faults, preventing cascading failures and enabling rapid corrective actions. This ensures the resilience and reliability of the grid, optimizing its operation even in the face of disruptions. The design of the deep learning approach comprises a long-term and short-term memory (LSTM) model. The LSTM model is trained on an electrical fault detection dataset that contains three-phase currents and voltages at one end serving as inputs and fault detection as outputs. The proposed LSTM model has attained an accuracy of 99.65 percent with an error rate of just 1.17 percent and outperforms neural network (NN) and convolutional neural network (CNN) models.

ViTR-Net: An unsupervised lightweight transformer network for cable surface defect detection and adaptive classification

As a crucial load-bearing component of the cable-stayed bridge, the cable requires surface defect detection to maintain its safety. The current deep-learning-based stay-cable surface defect detection still faces challenges, including insufficient actual defect samples, cumbersome labeling processes, and an effective method for further evaluation and classification of defects. To address these problems, an unsupervised vision transformer-based lightweight reconstruction network called ViTR-Net is proposed, which only uses normal stay-cable surface images to achieve end-to-end localization of defect areas in the image. ViTR-Net is a generative-based method that uses a hybrid network of convolutional neural networks and visual transformers as a feature extraction network. A feature correction fusion module is designed by integrating differential convolution and attention mechanisms to compensate for potential detail loss caused by feature extraction. Subsequently, a severity index for defects is established to assess the severity of various defects, and the detected defects are adaptively classified using the affinity propagation clustering algorithm. ViTR-Net achieves the highest defect detection accuracy and exquisite defect segmentation results on a realistic cable defect dataset through experimental validation. The proposed model can realize the surface defect assessment and adaptive classification of the stay cable, which has a good application prospect in the actual maintenance process of the stay cable.

Deep learning based fault detection of automobile dry clutch system using spectrogram plots

Dry friction clutches are extremely important in the context of power transmission systems. Continuous exposure to extreme heat and loading makes clutch extremely susceptible to various faults. The timely detection and diagnosis of such faults are of utmost importance to prevent any damage to internal components and also helps in avoiding transmission system failures. In this research study, a novel approach that leverages the power of transfer learning (a famous deep learning technique) is proposed to diagnose multiple types of clutch faults including, worn release fingers, fractured pressure plates, deteriorated pressure plates, loss of friction material and distorted tangential strips using spectrogram plots. To train and validate the diagnostic system, vibration readings were taken from a specially designed test rig with the help of piezoelectric accelerometer while the clutch system was operated under different load conditions of 0 (no load), 5 and 10 kg This procedure of data collection was then repeated to acquire the vibration data for all of the fault conditions by replacing the good with fault components individually. These vibration signals were further processed and transformed into spectrogram plots that serves as the input data for the deep learning models considered. Fine-tuning techniques were applied on pretrained networks to maximise the prediction accuracy of the models to effectively determine and diagnose faults in the clutch system. For this study 12 pre-trained networks were chosen namely, Xception, InceptionResNet, DenseNet, AlexNet, VGG16, GoogLeNet, VGG19, ResNet101, ResNet50, InceptionV3, MobileNetV2 and ShuffleNet. To optimize the performance of deep learning models, a systematic adjustment of hyperparameters such as the train-test split ratio, learning rate, optimizer and batch size for each network model was carried out. Through careful experimentation and analysis, significant improvements in fault classification accuracy were achieved thereby enhancing the reliability and effectiveness of the diagnostic system. From the results it was noted that 100% classification accuracy was displayed by AlexNet (for the no load condition and the 10 kg load condition) and GoogLeNet (for 5 kg load condition) with extremely low computation times.

Multi-scale quaternion CNN and BiGRU with cross self-attention feature fusion for fault diagnosis of bearing

In recent years, deep learning has led to significant advances in bearing fault diagnosis (FD). Most techniques aim to achieve greater accuracy. However, they are sensitive to noise and lack robustness, resulting in insufficient domain adaptation and anti-noise ability. The comparison of studies reveals that giving equal attention to all features does not differentiate their significance. In this work, we propose a novel FD model by integrating multi-scale quaternion convolutional neural network (MQCNN), bidirectional gated recurrent unit (BiGRU), and cross self-attention feature fusion (CSAFF). We have developed innovative designs in two modules, namely MQCNN and CSAFF. Firstly, MQCNN applies quaternion convolution to multi-scale architecture for the first time, aiming to extract the rich hidden features of the original signal from multiple scales. Then, the extracted multi-scale information is input into CSAFF for feature fusion, where CSAFF innovatively incorporates cross self-attention mechanism to enhance discriminative interaction representation within features. Finally, BiGRU captures temporal dependencies while a softmax layer is employed for fault classification, achieving accurate FD. To assess the efficacy of our approach, we experiment on three public datasets (CWRU, MFPT, and Ottawa) and compare it with other excellent methods. The results confirm its state-of-the-art, which the average accuracies can achieve up to 99.99%, 100%, and 99.21% on CWRU, MFPT, and Ottawa datasets. Moreover, we perform practical tests and ablation experiments to validate the efficacy and robustness of the proposed approach. Code is available at https://github.com/mubai011/MQCCAF.

A Deep Learning Method for Bearing Cross-Domain Fault Diagnostics Based on the Standard Envelope Spectrum.

Intelligent fault diagnostics based on deep learning provides a favorable guarantee for the reliable operation of equipment, but a trained deep learning model generally has low prediction accuracy in cross-domain diagnostics. To solve this problem, a deep learning fault diagnosis method based on the reconstructed envelope spectrum is proposed to improve the ability of rolling bearing cross-domain fault diagnostics in this paper. First, based on the envelope spectrum morphology of rolling bearing failures, a standard envelope spectrum is constructed that reveals the unique characteristics of different bearing health states and eliminates the differences between domains due to different bearing speeds and bearing models. Then, a fault diagnosis model was constructed using a convolutional neural network to learn features and complete fault classification. Finally, using two publicly available bearing data sets and one bearing data set obtained by self-experimentation, the proposed method is applied to the data of the fault diagnostics of rolling bearings under different rotational speeds and different bearing types. The experimental results show that, compared with some popular feature extraction methods, the proposed method can achieve high diagnostic accuracy with data at different rotational speeds and different bearing types, and it is an effective method for solving the problem with cross-domain fault diagnostics for rolling bearings.

Enhancing Power System Reliability Through Advanced Fault Diagnosis Methods: A Deep Learning Approach

In order to attain high reliability in power systems, effective fault detection techniques that can quickly identify and mitigate defects must be developed. In this article, we suggest a novel method for improving power system dependability by using deep learning techniques to identify faults. In order to achieve precise fault identification and classification, our technology uses deep neural networks to automatically learn and extract features from power system data. We thoroughly test our method on a range of fault scenarios and real-world datasets to determine its efficacy. The outcomes show how well our approach performs in comparison to conventional fault identification methods, underscoring the possibility of a major increase in power system reliability. By providing useful insights for improving problem diagnosis techniques in power engineering, this research advances system resilience and minimizing downtime.

Learning to Diagnose: Meta-Learning for Efficient Adaptation in Few-Shot AIOps Scenarios

With the advancement of technologies like 5G, cloud computing, and microservices, the complexity of network management systems and the variety of technical components have greatly increased. This rise in complexity has rendered traditional operations and maintenance methods inadequate for current monitoring and maintenance demands. Consequently, artificial intelligence for IT operations (AIOps), which harnesses AI and big data technologies, has emerged as a solution. AIOps plays a crucial role in enhancing service quality and customer satisfaction, boosting engineering productivity, and reducing operational costs. This article delves into the primary tasks involved in AIOps, such as anomaly detection, and log fault analysis and classification. A significant challenge identified in many AIOps tasks is the scarcity of fault sample data, indicating a natural alignment of these tasks with few-shot learning. Inspired by model-agnostic meta-learning (MAML), we propose a new anomaly detector, MAML-KAD, for application in various AIOps tasks. Observations confirm that meta-learning algorithms effectively enhance AIOps tasks, showcasing the wide-ranging application prospects of meta-learning algorithms in the field of AIOps. Moreover, we introduced an AIOps platform that embeds meta-learning within its diagnostic core and features streamlined log collection, caching, and alerting to automate the AIOps workflow.

CNN-based defect detection and classification of PV cells by infrared thermography method

ABSTRACT As a crucial component of the photovoltaic (PV) power generation system, effective defect detection of PV cells can help to ensure the power generation efficiency and the system safety. The photo-thermal-radiation infrared thermography experimental system has been built, and the infrared images of normal and defective PV cells were collected by an IR camera. An effective convolutional neural network (CNN) based model with the residual architecture is designed here to detect and classify the defects. An offline date augmentation method is performed to overcome the insufficient image dataset in the step of dataset establishment, so as to improve the defects classification capability of the designed CNN model. The defects classification accuracy of the proposed model is obtained as 96.96% with 3-class. In order to verify the defect classification ability of the designed CNN model, other four models were adopted for comparative analysis. The results indicate that the proposed CNN model has advantages in defect classification of PV cells. The completed work can provide meaningful exploration in the industrial application of defect detection of PV cells.

Multi‐view synergistic enhanced fault recording data for transmission line fault classification

AbstractFault recorded data has been proven to be effective for fault diagnosis of overhead transmission lines. Utilizing deep learning to mine potential fault patterns in fault recording data is an inevitable trend. However, it is usually difficult to obtain massive labeled fault recording data, which results in deep learning‐based fault diagnosis models not being adequately trained. Although data augmentation methods provide ideas for expanding the training data, existing data augmentation algorithms (e.g. random perturbation‐based augmentation) may lead to distortion of multi‐view data, that is, time domain data and frequency domain data of the fault recorded data, which results in the inconsistency of physical properties and statistical distributions of the generated data and the actual recording data, and misguides the training of the models. Hence, this study proposes a transmission line fault classification method via the multi‐view synergistic enhancement of fault recording data. The methodology proposes to start with a synergistic enhancement of multi‐view data such as time and frequency domains of fault recording data, and utilizes contrastive learning to further improve the performance of the fault classification model while ensuring that the generated data is not distorted. Experimental results on three real‐world datasets validate the effectiveness of the proposed method.

Learning-Based Approach for Automated Surface Inspection with Industrial Tomography Imaging

Abstract In recent years, advanced deep learning techniques have emerged as pivotal tools in enabling the development of robust vision-based solutions for steel surface inspection. This resulted in enhanced inspection accuracy, all while significantly reducing costs in the manufacturing industry. However, the lack of actual steel surface defects datasets currently places a certain constraint on further research into classifying those anomalies. As a consequence, the Convolutional Neural Network (CNN) technique, known for its prowess in image-related tasks, faces certain challenges, especially in classifying less common defects. This work proposes a novel hybrid CNN model with a Support Vector Machine (SVM) classifier at the output layer for surface defects classification. The features extracted from the pre-trained ResNet152 and EfficientB0 CNN algorithms are concatenated and fed to the SVM layer for classification. Extensive experiments on a merged dataset consisting of the publicly available Northeastern University (NEU) dataset and Xsteel surface defect dataset (X-SDD) are carried out and the accuracy and F1 scores are calculated for performance evaluation. The merged dataset contains eleven typical defect types with a total of 2660 defect images. Then, the adopted algorithm is compared with ten fine-tuned deep learning models to evaluate the performance of transfer learning for steel defect detection and identification. The evaluation results show that the deep feature extraction and SVM classification produced better results than the transfer learning. Finally, the proposed classifier model is validated on a newly collected dataset from a Computed Tomography scanner with an accuracy reaching over 96%.

Hot rolled steel surface defect detection and classification using an automatic ensemble approach

This study introduces an ensemble-based Deep Neural Network (DNN) model for detecting defects on steel surfaces. The method suggested in this study classifies steel surface conditions into six possible fault categories, namely, crazing, inclusion, rolled in, pitted surface, scratches, and patches. The images undergo preprocessing and extraction of features in spatial and frequency domains using image segmentation techniques such as grey level difference method (GLDM), fast Fourier Transform (FFT), grey level co-occurrence matrix (GLCM), texture analysis and discrete wavelet transform (DWT). The ensembling of image features into a fused feature pool is carried out after the preprocessing of input images that are provided as input to a light-weight neural network model for training and testing. The performance of the model is comprehensively evaluated via an ablation study both before and after ensembling. In addition, the model capability is effectively analyzed using receiver operating characteristics (ROC) curve, confusion matrix from which classification accuracy of the model could be obtained and other parameters including precision and f1-score. It was observed that the proposed deep learning network presents phenomenally high accuracy of 99.72% for detection and classification of steel surface faults. This result was found to be superior when compared with the performance of the same neural network over each feature type individually. This study also compares the classification results of the model built based on the ensembled feature set with the results of various other classification approaches available in literature. The ensemble-based model could potentially be integrated into existing inspection systems for real-time, efficient and robust condition monitoring of steel surfaces.