Related references
Note: Only part of the references are listed.
Article
Computer Science, Artificial Intelligence
Xiaojun Chang et al.
Summary: Scene graph is a structured representation of a scene, expressing objects, attributes, and relationships. With the development of computer vision, people aim for a higher level of understanding and reasoning about visual scenes. Scene graphs have attracted researchers' attention as a powerful tool for scene understanding.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Chuanjiang Li et al.
Summary: Deep learning-based fault diagnosis methods have achieved remarkable progress, but they are often coarse grained and require large amounts of data, which cannot identify the root causes of mechanical system failures at a finer granularity with limited fault data. In this study, a novel attention-based deep meta-transfer learning (ADMTL) method is proposed to address the challenges of fine-grained fault feature extraction and limited model generalization ability. The proposed method achieves excellent performance in few-shot fine-grained fault diagnosis tasks.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Automation & Control Systems
Cunxiao Shen et al.
Summary: This paper proposes a novel data-driven self-attention based deep learning model for chiller fault diagnosis. It addresses the data imbalance problem using stable synthetic minority oversampling technique (SSMOTE) and achieves accurate diagnosis through the combination of self-attention mechanism and traditional residual blocks.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Automation & Control Systems
Tianci Zhang et al.
Summary: Data-driven intelligent diagnosis models aim to extract health information from massive monitoring data, but the limited size of faulty monitoring data in engineering scenarios has made few-shot fault diagnosis an important research point. This study proposes a prior knowledge-augmented self-supervised feature learning framework to reduce the amount of training data required by integrating prior diagnosis knowledge.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Automation & Control Systems
Saibo Xing et al.
Summary: This article proposes an adaptive knowledge transfer with multiclassifier ensemble (AKTME) method based on the paradigm of continual machine learning for addressing the few-shot diagnosis problem in machine faults. AKTME represents knowledge using learnable filter kernels (FKs) and transfers knowledge through a technique called continual weighted updating (CWU). Experimental results show that AKTME achieves higher diagnosis accuracies compared to recently proposed methods and tends to improve accuracy through pretraining on more auxiliary tasks.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Automation & Control Systems
Haidong Shao et al.
Summary: In this article, a modified stacked autoencoder (MSAE) that uses adaptive Morlet wavelet is proposed for automatically diagnosing various fault types and severities of rotating machinery. Experimental results show that the proposed method is superior to other state-of-the-art methods.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2022)
Article
Computer Science, Artificial Intelligence
Yong Feng et al.
Summary: The recent advances in intelligent fault diagnosis demonstrate the strong capabilities of deep learning in automatic feature extraction and accurate identification of fault signals, yet challenges such as data scarcity and varying working conditions may impact model performance. The tool proposed to address these challenges is meta-learning, which quickly adapts to new tasks using small samples and has great potential in few-shot and cross-domain fault diagnosis. The lack of a survey to summarize existing work and look into the future is noted, with this paper comprehensively investigating deep meta-learning in fault diagnosis from three perspectives.
KNOWLEDGE-BASED SYSTEMS
(2022)
Review
Automation & Control Systems
Tianci Zhang et al.
Summary: Research on intelligent fault diagnosis using artificial intelligence technologies has achieved significant progress, particularly in the field of S&I-IFD. Existing strategies include data augmentation, feature learning, and classifier design. Future research directions involve meta-learning and zero-shot learning.
Article
Automation & Control Systems
Yong Feng et al.
Summary: In this paper, a semi-supervised meta-learning network with attention mechanism is proposed for few-shot fault diagnosis in mechanical systems. The method utilizes unlabeled data to improve fault recognition and achieves outstanding adaptability in different situations, as demonstrated through experiments with bearing vibration datasets.
Article
Automation & Control Systems
Zhenzhen Jin et al.
Summary: In this paper, a weak fault diagnosis method for train axle box bearings is proposed based on parameter optimization Variational Mode Decomposition (VMD) and improved Deep Belief Network (DBN). By optimizing algorithm parameters and extracting fault feature information, the diagnostic accuracy of the bearings can be improved.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Automation & Control Systems
Bing Li et al.
Summary: This study introduces a transfer learning based extreme learning machine for fault diagnosis of a turbofan engine. The method aligns the distribution discrepancies of the data, learns transferable cross domain features, and preserves the properties and structures of the source domain. Extensive experiments demonstrate that the proposed method achieves high diagnostic accuracy.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Engineering, Multidisciplinary
Yanxin Wang et al.
Summary: This article proposes a data-driven artificial intelligence method for high-voltage circuit breaker fault diagnosis. By introducing an attention mechanism and few-shot transfer learning, the proposed method achieves high accuracy and robustness in fault diagnosis with limited samples.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Caixia Yan et al.
Summary: This paper proposes a method called ZeroNAS that integrates neural architecture search (NAS) techniques into zero-shot learning (ZSL). It uses adversarial training to search for desirable architectures in a specially designed search space for GANs. Extensive experiments show that ZeroNAS outperforms state-of-the-art ZSL and GZSL approaches.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Automation & Control Systems
Yiming Xiao et al.
Summary: This article proposes a novel joint transfer network for unsupervised bearing fault diagnosis from the simulation domain to the experimental domain. It uses bearing simulation data to construct the source domain, an improved loss function to achieve alignment across domains, and a weight allocation mechanism to suppress negative transfer.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2022)
Article
Computer Science, Artificial Intelligence
Rui Wang et al.
Summary: This study proposes a novel federated adversarial domain generalization network for machinery fault diagnosis, which aims to build a global fault diagnosis model for multiple clients under data privacy conditions. Experimental results show that the proposed network can achieve significant performance improvement for diagnosis methods with federated learning.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Engineering, Multidisciplinary
Xiao Zhang et al.
Summary: The article introduces a few-shot fault diagnosis method under open-set domain adaptation scenario, which calculates prototypes based on feature distance and enhances the compactness of samples, achieving accurate classification and effective rejection.
Article
Engineering, Electrical & Electronic
Jiafeng Li et al.
Summary: In this article, a command filter-based adaptive fuzzy finite-time output feedback control method is investigated for the electrohydraulic servo system. Uncertainties in the system are estimated using fuzzy logic systems (FLSs), and a fuzzy state observer is established to approximate unmeasurable parameters. A command filter-based control method is proposed to achieve high tracking precision. Simulation results demonstrate the effectiveness of the proposed control.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Engineering, Electrical & Electronic
Xue Wang et al.
Summary: This article proposes a decoupling network-based method for infrared and visible image fusion, which effectively retains the texture details and luminance information of the source images while preserving the high-contrast regions. Extensive experiments demonstrate that the proposed method generates fused images with saliency objects and clear details, outperforming other state-of-the-art methods.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Computer Science, Artificial Intelligence
Jia Luo et al.
Summary: The study introduces an imbalanced fault diagnosis method based on conditional-deep convolutional generative adversarial network (C-DCGAN), which improves fault diagnosis accuracy and generalization ability by balancing the dataset using generated samples.
JOURNAL OF INTELLIGENT MANUFACTURING
(2021)
Article
Automation & Control Systems
Rachna Vaish et al.
Summary: The study focuses on providing a comprehensive review of machine learning-based power system fault diagnosis, addressing the issues in conventional fault diagnosis and presenting a framework for machine learning-based fault diagnosis. Various unsupervised and supervised learning techniques are discussed, along with a brief overview of reinforcement learning and transfer learning in the context of power system fault diagnosis. Additionally, the research trends, key issues, and directions for future research are highlighted.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Automation & Control Systems
Sebastian Schwendemann et al.
Summary: This study introduces a novel technique based on an intermediate domain that achieves improved accuracy on small labeled datasets, and proposes a Convolutional Neural Network architecture for transfer learning along with a new transfer learning approach. The method outperforms traditional combinations on all datasets for both unsupervised and semi-supervised learning, and shows promising results in handling different types of bearings and varying rotation speeds.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Automation & Control Systems
Qun-Xiong Zhu et al.
Summary: A new virtual sample generation approach CGAN-VSG is proposed to address the issue of insufficient data in data-driven soft sensing modeling of industrial processes. By integrating LOF with K-means++ algorithm and utilizing CGAN to generate virtual input and output samples, the proposed method enhances the accuracy of soft sensors with small data. Simulation results demonstrate the superiority of CGAN-VSG in terms of accuracy compared to other state-of-the-art methods.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Automation & Control Systems
Tianci Zhang et al.
Summary: This study proposes an intelligent fault diagnosis method based on generative adversarial networks, where the generator generates mechanical signals to expand the training dataset and the classifier has the ability to extract fault features and classify with high accuracy under small sample conditions.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Article
Computer Science, Artificial Intelligence
Cunjun Wang et al.
Summary: This study introduces an intelligent model for few-shot fault diagnosis, which utilizes similarities of sample pairs to classify samples in order to enhance classification performance. Results demonstrate the superior performance of the proposed model in bearing fault diagnosis.
Article
Engineering, Industrial
Min Xia et al.
Summary: The intelligent fault diagnosis framework based on DT and deep transfer learning enables accurate machine fault diagnosis with limited measured data. This method outperforms other state-of-the-art data-driven methods in achieving intelligent fault diagnosis.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Engineering, Multidisciplinary
Baokun Han et al.
Summary: This study introduces a novel transfer learning method called hybrid distance-guided adversarial network (HDAN) to address the issue of degradation in diagnostic accuracy caused by data shift in intelligent fault diagnosis. Experimental results demonstrate that HDAN can achieve better feature cluster performance of the same class in different domains.
Article
Automation & Control Systems
Qunwang Yao et al.
Summary: This paper proposes a Multiscale Domain Adaption (MSDA) method to improve classical Deep Domain Adaption (DDA) models by addressing their inherent drawbacks. MSDA, with its MSDA block constructed based on multiscale convolution networks, can extract more domain-invariant features and simplify the calculation of domain confusion losses, thereby enhancing the domain adaption abilities of DDA models.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Automation & Control Systems
Chuanxia Jian et al.
Summary: The study proposes a fault diagnosis method based on active and semi-supervised learning, which improves model performance by selecting uncertain unlabelled samples and using heterogeneous classifiers, suitable for fault diagnosis with a small training set.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)
Article
Engineering, Multidisciplinary
Rugen Wang et al.
Summary: Fault diagnosis is crucial for the stability and reliability of mechanical equipment operation, and recent studies have shown that deep learning outperforms traditional methods in this field. However, the issue of sample imbalance can hinder deep learning models in training and affect classification accuracy. To address this problem, an enhanced generative adversarial network (E-GAN) is proposed, which utilizes additional generated samples to balance the training set and integrates a clustering algorithm for fault classification. Experiment results demonstrate that E-GAN significantly improves classification accuracy compared to other methods.
Article
Engineering, Mechanical
Bin Yang et al.
Summary: This paper introduces a deep partial transfer learning network (DPTL-Net) that can selectively transfer diagnostic knowledge across asymmetric domains, demonstrating better performance in two case studies.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Jingyi Xu et al.
Summary: The study introduces a feature disentanglement framework to augment samples in few-shot classification while preserving their class-discriminative features. By separating the feature representation into intra-class variance and class-discriminative information, the framework effectively enhances features.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Article
Engineering, Electrical & Electronic
Tongyang Pan et al.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2020)
Article
Computer Science, Artificial Intelligence
Bo Zhao et al.
KNOWLEDGE-BASED SYSTEMS
(2020)
Article
Automation & Control Systems
Jianbo Yu et al.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2020)
Article
Engineering, Mechanical
Yongtao Hu et al.
CHINESE JOURNAL OF MECHANICAL ENGINEERING
(2019)
Article
Computer Science, Artificial Intelligence
Laurens van der Maaten et al.