Related references
Note: Only part of the references are listed.
Article
Computer Science, Information Systems
Bilgin Umut Deveci et al.
Summary: Bearings are critical components in rotating machinery, and undetected faults can result in financial and human losses. This study addresses the challenge of choosing the most suitable 2-D imaging method and neural network for early detection of bearing faults. By training eighteen imaging methods with four different networks using the same vibration data and training metrics, the study identifies Scattergram Filter Bank 1 as the best image input and ResNet-50 as the most effective network, achieving an accuracy of 99.89%. Prior to this research, Scattergram images have never been used for bearing fault classification. Ten out of 72 methods used in this study achieved accuracies higher than 99.5%.
INFORMATION SYSTEMS FRONTIERS
(2023)
Article
Chemistry, Analytical
Yubin Yoo et al.
Summary: Bearing defects can cause unexpected downtime, costly repairs, and safety hazards in rotating machines and equipment. Deep learning models have shown promise in diagnosing these defects, but their complexity often leads to high computational costs. This paper proposes a new approach that reduces input data dimensionality and optimizes model structure simultaneously. By downsampling vibration sensor signals and constructing spectrograms, a low-dimensional input data was achieved. The experimental results show that this method is highly efficient in computation and maintains outstanding classification performance.
Article
Multidisciplinary Sciences
Lu Xiao et al.
Summary: Bearings are vital components in mechanical equipment, and detecting their failures is crucial for ensuring the equipment's proper functioning and preventing accidents. To address this issue, we propose a graph neural network-based method for bearing fault detection. This method constructs a graph based on sample similarity, maps the graph using a graph neural network for feature extraction, and then detects faults using a base detector.
SCIENTIFIC REPORTS
(2023)
Article
Engineering, Chemical
Guoguo Wu et al.
Summary: Rolling element bearings (REBs) are the most common cause of machine breakdowns. Traditional fault diagnosis methods rely on feature extraction and signal processing, which can be affected by the complexity of patterns and the need for expert knowledge. This paper proposes a novel signal-to-image method using continuous wavelet transform (CWT), which enhances feature extraction and eliminates the need for manual extraction.
Review
Mathematics, Interdisciplinary Applications
Mohammad Mustafa Taye
Summary: Convolutional neural networks (CNNs) are widely used in image recognition and classification, with applications ranging from object recognition to face recognition. CNNs learn a hierarchy of features from input images through a process of convolution, enabling them to extract complex features that are invariant to distortion and translation. This study aims to identify research gaps and provide detailed insights into the building blocks and important issues of CNNs.
Article
Engineering, Manufacturing
Ji-Won Jin et al.
Summary: The automotive wheel bearing is a crucial part of cars that transfers rotation and supports the vehicle's weight. Its lifespan can be categorized into the raceway and flange lives, with the former being determined analytically using the basic rating life calculation method and the latter being predicted through fatigue analysis. This study examined the fatigue analysis of the automotive wheel bearing flange while considering rotation, and verified the results by comparing them with test results.
INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING
(2023)
Article
Engineering, Electrical & Electronic
Changpu Yang et al.
Summary: This article studies the benefits of noise in intelligent fault diagnosis and proposes a method to improve classification accuracy by injecting moderate noise. Experimental results demonstrate the effectiveness of this method.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Engineering, Multidisciplinary
Jun Li et al.
Summary: The paper introduces a novel fault diagnosis approach for rolling bearings, combining DA-RNN and CBAM technologies, which enhances diagnostic accuracy by handling imbalanced datasets and utilizing convolutional neural networks.
Article
Computer Science, Information Systems
Selen Ayas et al.
Summary: This paper presents a novel deep learning-based model for fault detection and classification of motor bearing. Time domain signals are converted to grayscale images using a proposed signal-to-image conversion method, and a deep residual learning network is utilized to learn the mapping between images and the health condition of the motor bearing. Experimental results on a commonly used vibration dataset show that the proposed model outperforms knowledge-based methods with an average accuracy of 99.98%.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Engineering, Mechanical
P. Akhenia et al.
Summary: Condition monitoring and diagnosis of bearings are crucial for the safety of rotating machines, and selecting suitable signal processing techniques for feature vector construction is a challenge. By utilizing deep learning algorithms and data augmentation techniques, the proposed method can significantly improve the accuracy of fault severity detection.
PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART C-JOURNAL OF MECHANICAL ENGINEERING SCIENCE
(2022)
Article
Engineering, Mechanical
Jie Huan et al.
Summary: This paper analyzes the oil film thickness and capacitance in bearings and explores the relationship between oil film thickness, rotational speed, and load. The capacitance of the bearing is affected by the Hertz contact area, oil film thickness, and lubricant quality. Orthogonal tests are conducted to study the influence of voltage, rotation speed, and load on bearing operation. The results show that voltage, load, and rotation speed are the main factors affecting raceway damage width, surface roughness, and vibration amplitude. The use of mixed ceramic bearings and different types of grease is explored to solve the problem of shaft current, and the advantages and disadvantages of each combination are analyzed and compared. The results show that non-conductive grease reduces damage, surface roughness, and vibration better than conductive grease.
PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART C-JOURNAL OF MECHANICAL ENGINEERING SCIENCE
(2022)
Article
Chemistry, Analytical
Matthias Kahr et al.
Summary: This study conducts 3D multi-body simulations to create synthetic training data for deep learning convolutional neural networks, aiming to address challenging tasks such as high quality data gathering and historical data limitations. By superimposing measurement data with noise and converting it into 2D images for training, the algorithm's performance is validated using measurements of damaged bearings.
Article
Computer Science, Artificial Intelligence
Happy Nkanta Monday et al.
Summary: This study proposes a modified deep learning approach that combines wavelet convolutional capsule network and enhanced super resolution generative adversarial network for fault diagnosis and classification. By applying continuous wavelet transform and super resolution techniques, the model achieves high accuracy in diagnosing faults using 2D time-frequency images. Experimental results demonstrate that the proposed method outperforms traditional deep learning models in diagnosing motor bearing faults.
COMPLEX & INTELLIGENT SYSTEMS
(2022)
Article
Computer Science, Information Systems
Xiaoping Zhao et al.
Summary: This study proposes a small-sample bearing fault diagnosis method based on an improved Siamese neural network, achieving higher fault diagnosis accuracy and better generalization in the case of small samples.
JOURNAL OF CLOUD COMPUTING-ADVANCES SYSTEMS AND APPLICATIONS
(2022)
Article
Chemistry, Analytical
Jialin Yan et al.
Summary: This paper proposes a rolling-bearing fault diagnosis model MTF-ResNet based on a Markov transition field and deep residual network. By using a sliding window to augment the raw vibration signal data and converting vibration signal samples into two-dimensional images, the model performs feature extraction through a deep residual neural network in image classification to identify the severity and location of bearing faults. Experimental results show that MTF-ResNet model has better average accuracy compared to other diagnostic methods.
Article
Chemistry, Multidisciplinary
Qingbin Tong et al.
Summary: This paper proposes an auxiliary classifier generative adversarial network with spectral normalization (ACGAN-SN). By generating fake data, using label constraints, and spectral normalization constraints, it improves the performance of the fault diagnosis model under small and imbalanced fault samples.
APPLIED SCIENCES-BASEL
(2022)
Proceedings Paper
Materials Science, Multidisciplinary
Ekta Yadav et al.
Summary: This paper provides a review of the importance of bearings in engines and machines, discussing their role, factors affecting performance, material selection criteria, and defect detection techniques.
MATERIALS TODAY-PROCEEDINGS
(2022)
Article
Chemistry, Analytical
Hongtao Tang et al.
Summary: An intelligent fault diagnosis strategy for rolling bearings based on grayscale image transformation, a generative adversative network, and a convolutional neural network was proposed to solve the challenge of achieving highly accurate rolling bearing fault diagnosis. Through experiments, it was demonstrated that this strategy can reach more than 99.6% recognition accuracy even under substantial environmental noise interference or changing working conditions, providing good stability in the presence of a severe imbalance in fault data.
Proceedings Paper
Computer Science, Artificial Intelligence
Duo Li et al.
Summary: The study introduces a new operation named involution to replace standard convolution for vision tasks, showing improved performance of models while reducing computational costs.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Article
Engineering, Electrical & Electronic
Zhenxiang Li et al.
Summary: The ACWGAN-GP model proposed in this article addresses the issue of data imbalance in fault diagnosis by generating high-quality samples for minority classes, gradually enhancing the imbalanced data set, and effectively improving the accuracy of fault diagnosis. This model also outperforms other widely used methods in sample generation.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Computer Science, Artificial Intelligence
Bo Zhao et al.
KNOWLEDGE-BASED SYSTEMS
(2020)
Article
Automation & Control Systems
Sheng Guo et al.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2020)
Article
Computer Science, Interdisciplinary Applications
Huaqing Wang et al.
COMPUTERS IN INDUSTRY
(2019)
Article
Computer Science, Artificial Intelligence
Te Han et al.
KNOWLEDGE-BASED SYSTEMS
(2019)
Article
Chemistry, Analytical
Hongmei Li et al.
Article
Automation & Control Systems
Qi Xuan et al.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2019)
Article
Multidisciplinary Sciences
Hongchuan Cheng et al.
SN APPLIED SCIENCES
(2019)
Article
Computer Science, Information Systems
Wentao Mao et al.
Article
Computer Science, Information Systems
Ruyi Huang et al.
Article
Engineering, Mechanical
Attila Gonda et al.
Article
Computer Science, Information Systems
Huihui Qiao et al.
Review
Computer Science, Information Systems
Ajay Shrestha et al.
Article
Automation & Control Systems
Long Wen et al.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2018)
Article
Engineering, Multidisciplinary
Thibaud Plazenet et al.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2018)
Article
Computer Science, Artificial Intelligence
Zirui Wang et al.
Article
Engineering, Mechanical
Christian Wagner et al.
Article
Engineering, Multidisciplinary
Aleksei Romanenko et al.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2016)
News Item
Biochemical Research Methods
Nicole Rusk
Article
Engineering, Mechanical
Seung Pyo Lee et al.
TRANSACTIONS OF THE KOREAN SOCIETY OF MECHANICAL ENGINEERS A
(2012)
